Ground Loops and other High Voltage Issues

This forum is for specialized infomation important to the construction and safe operation of the high voltage electrical supplies and related circuitry needed for fusor operation.
Post Reply
User avatar
Dennis P Brown
Posts: 2027
Joined: Sun May 20, 2012 2:46 pm
Real name: Dennis P Brown
Location: Glen Arm, MD

Ground Loops and other High Voltage Issues

Post by Dennis P Brown » Mon Mar 04, 2013 11:50 pm

Got two rather shocking experiences today – the first was indeed a high voltage shock from a very small cap/diode ladder system to create a high volt multiplier (25 kV; supplied from a 12 volt battery.) Then later, after I decided to ground the unit’s case rather than float it, I then discovered that ground loops can go many strange places.

My main ground is a metal water pipe (that is fed from a well and also is the main ground for my house panel.) When I fired up the HV supply again, a light on a different circuit had a large, loud and continuous arc jumping across part of its bulb. Apparently, the high voltage ran through the water to my vacuum system’s case (this has water cooling lines in it), then jumped to an electrical part’s neutral wire and then ran back to the outlet it was connected to; only then to find the other circuit’s ground in order to reach the light bulb! Now that was convoluted.

Bottom line: even a true ground is not always trust worthy and high voltage can find a way … . Believe this is two lessons learned by me: never trust any high voltage system (even commercial) and always ground everything.

While I am still confused on why the high voltage multiplier (standard cap/diode ladder fed by a 2 kV power supply) would feed back into the case of the power supply, still it does and I'd better work around that issue.

I guess part (all) the reason is that the case and bottom of the ladder share a common ground but I thought that did not get charged …boy, was that an assumption that bite me; rather than 2 kV that was the full 25 KV; since the caps are toys, no harm but did learn a painful lesson.

The reason I was messing with this supply is that this will be a positive 25 kV, low current spray for my Van de Graaf belt. I’m trying to finish the high voltage field supply for my accelerator. About the last item to complete (didn’t realize how many sub systems that required building for a commercial unit!) So, I had to confirm its voltage and polarity because the ESLA won’t work if those are not correct. While this is the last major job, still have put off a few minor issues that have also bit me, but in a different, less painful, manner.

I guess the old saying is true: there is the right way, the wrong way, and the way that causes the most work for the least results … wait, I just made that up. Must be the shock still firing a few random neurons … .

Tyler Christensen
Site Admin
Posts: 551
Joined: Tue Feb 24, 2009 2:08 am
Real name:

Re: Ground Loops and other High Voltage Issues

Post by Tyler Christensen » Mon Mar 04, 2013 11:54 pm

There isn't enough information here for a full diagnosis, but what happened shouldn't have happened if everything is assembled properly. Something is very dangerously wrong with your setup. The fact that a grounded chassis was at a high potential implies, quite simply, that it's not actually grounded. I'd strongly recommend thoroughly confirming every ground, both in your house as well as in your high voltage wiring. Something just doesn't sound right here.

User avatar
Dennis P Brown
Posts: 2027
Joined: Sun May 20, 2012 2:46 pm
Real name: Dennis P Brown
Location: Glen Arm, MD

Re: Ground Loops and other High Voltage Issues

Post by Dennis P Brown » Tue Mar 05, 2013 12:32 am

I agree with you; and I have a suspicion on the HV case problem. I am very confused on the water ground since the water was running and the pipe is metal.

User avatar
Chris Bradley
Posts: 2930
Joined: Fri May 02, 2008 11:05 am
Real name:

Re: Ground Loops and other High Voltage Issues

Post by Chris Bradley » Tue Mar 05, 2013 12:48 am

A stack of voltage stages has a capacitance from top of the stack to the bottom.

Say the top of the stack is at -20kV and it sparks to ground. All of a sudden the top is now at zero, which means that the bottom, that had been at ~~0V, now jumps to +20kV because it is capacitively coupled and because there was some 20kV or so across the caps. As the top discharges, it will pull the bottom up.

The driving circuit/transformer that is attached to that is now pulled through that voltage suddenly. The HV will then end up anywhere that the driver circuit is attached to.

This is how driver transformers/circuits at the bottom of a stack tend to get fried during HV arcing events at the top of it.

This may or may not apply in your situation. As mentioned by Tyler, insufficient information. But where a stack is arcing you need to look out for this effect. The fix is either to drive the stack with a charge pump arrangement with some TVS protection built into it, or to use resistors so the impedance for the driver is lower to ground than it is to the stack by a few oom.

User avatar
Dennis P Brown
Posts: 2027
Joined: Sun May 20, 2012 2:46 pm
Real name: Dennis P Brown
Location: Glen Arm, MD

Re: Ground Loops and other High Voltage Issues

Post by Dennis P Brown » Wed Mar 06, 2013 11:00 am

After checking the system, I have found the main fault that caused the dangerous shocking problem for my Van de Graaf high voltage spray system - the ground for my high voltage ladder multiplier (a diode/cap system) was wired to the case ground of the HV power supply (this is operated by a 12 volt battery). I did not think through what I was doing - which is very often the bases for all major HV mistakes ... .

Normally, all my electronics metal support structures are connected to my master 'fail-safe' ground (the water pipe.) I keep this system separate from the universal 'center' ground found in all the outlets. Foolishly, since I was wiring the various parts together in a rush manner (major error), I didn't think about how I was using my grounds. The fact that the small HV power supply uses a battery was what gave me a problem; that type of supply has no "center grounding plug".

So, I used this master 'fail safe' ground for both the HV power supply case and the 'base ground' for the voltage multiplier ladder (diode/cap system.) While using the master fail safe ground for the HV case is acceptable, this ground should never have been used for grounding the base of the HV ladder; that part of my circuit should have been connected to the standard center ground of my home outlet.

This grounding setup was stupid and I paid a painful lesson.

Thanks for the advice - as always (for me, at least) this advice helped me to rethink this issue more clearly and aidded me in creating a safer system. I will be far more careful in the future with my HV supplies!

User avatar
Chris Bradley
Posts: 2930
Joined: Fri May 02, 2008 11:05 am
Real name:

Re: Ground Loops and other High Voltage Issues

Post by Chris Bradley » Wed Mar 06, 2013 2:59 pm

Dennis P Brown wrote:
> So, I used this master 'fail safe' ground for both the HV power supply case and the 'base ground' for the voltage multiplier ladder (diode/cap system.) While using the master fail safe ground for the HV case is acceptable, this ground should never have been used for grounding the base of the HV ladder; that part of my circuit should have been connected to the standard center ground of my home outlet.
This sounds all pretty confusing to me, and that might mean you are pretty confused by it too.

There should only be one 'ground' to all your interconnecting circuits, and it should be the shortest and 'fattest' connection that the setup permits, and usually one picks the 'biggest and ugliest' bit of metal in the system (viz. the fusor shell) so that low impedance connections can be made to it with big solid connections, e.g. big bolts and ring crimps. Short-and-fat connections mean the lowest inductance path, which means low differential voltages where very short duration HV pulses propagate in the system.

One then, typically 'earths' that ground point of such a setup, which will stop the ground bouncing around with respect to earth. That earth connection might simply be the earth line of a regular socket. There should be no particular need to connect to water pipes or such things.

So, talk of 'master fail safe grounds' and such do make me wonder what we are talking about here, exactly. It sounds like you might be creating a return path that loops down around the water pipes and back via the domestic electrical earth. Though I am not particularly familiar with US domestic wiring systems so I'm not sure what that involves, exactly, this would seem to me to be 'a wrong path' for any residual currents to try to take in any sort of domestic circuit.

If you want to draw out a circuit of what you think you had and why you say it was wrong, there maybe some useful learning that might come out of discussing it.

User avatar
Dennis P Brown
Posts: 2027
Joined: Sun May 20, 2012 2:46 pm
Real name: Dennis P Brown
Location: Glen Arm, MD

Re: Ground Loops and other High Voltage Issues

Post by Dennis P Brown » Wed Mar 06, 2013 6:54 pm

Yes, Chris, I believe that the complete ground system that I am using might 'loop' around. What I am calling a master ground is a wire that is connected to a metal water pipe that, in turn, connects to an underground water well. I use this to ground my main metal frame that all my electronics, vacuum system, and accelerator are built into.

My house electral power system uses the standard American three prong plug where, on the plug, one prong is a ground and the other two are the 'hot' (115 volt, 60 Hz) and neutral. This is a standard plug arrangement that uses the ground prong on this plug to protect electronic cases as a default 'path to ground'.

Most of my electrical equipment has std. three prong plugs (the power pannel's hot, neutral and ground) that connect to the wall outlet (all have been confirmed as working correctly with valid neutrals and grounds.)

Sorry I have not managed to explain this in a fashion that anyone could figure out.

User avatar
Richard Hull
Moderator
Posts: 12496
Joined: Fri Jun 15, 2001 1:44 pm
Real name: Richard Hull

Re: Ground Loops and other High Voltage Issues

Post by Richard Hull » Thu Mar 07, 2013 2:55 pm

In th' biz, a "star ground" is the norm.

Take a point that you know is ground; put a 1/4X20 bolt, stud lug through a metal panel and make all connections to all gear from this one point. Do not ground serially! (unit to unit) Ground individual units to star point lug.

In good design, all items needed to be grounded within the system are located as close together as possible such that all connections to the star point are short. If this is not possible, locate your gear suitably and then establish your star ground lug at some central point amongst the gear.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
Retired now...Doing only what I want and not what I should...every day is a saturday.

Post Reply