Voltage spikes on ground wire.

[Rod]

I believe when coils drop out is when the spikes occur. The collapsing field creates the spike.

Rod (the CNC dude)

 

[sirhiss2]

Maybe I'm way off here, but have you checked for possible ground loops? I know that with various instrumentation cables as well as control cables they should only be bonded at one end. Long runs of cable will produce weird stuff if not done this way. I would also agree with Mordred, Harmonics could be a problem especially if there are alot of drives around. Fluke has a really good meter for analyzing this. I can't remember the Part number right now. When I get back to my office I will post it.

 

[Bruce99]

Bonding

Grounding and bonding fun!!!
Make sure that the cable coming off the amp is grounded on one end only. If it is running a long distance you may get a voltage between the two grounded systems. I would also "float" the 24 volt supply.
Remove the DC common from ground and mount the supply on insulated stand-offs. Keep the DC separate or ungrounded.
You should never get a voltage at those points if your system was properly bonded to ground. Ive seen MCCs and transformers in plants "grounded" to the building steel. This is OK only if the steel is bonded by copper conductor to the plant system ground grid. The steel alone is not considered a bonding conductor. Ive just finished a job where 30 volts was measured to PDC ground from the VFD grounds in a large MCC. (no one thought to measure this because it was grounded to the steel). It took 15 feet of 000 welding cable to fix a problem that cost thousands in down time.
There are lots of good ideas on this thread.
Floating the DC supply is a lot of work, and you may not want to get into all the wiring. Check the grounds and see if there is a copper ground cable running to a grid somewhere.
If your amp is running a signal to another building, make sure the two buildings have a common ground grid or you could see a potential between the two points. Telephone cable grounds also have to be bonded as well.
Take care, Bruce

 

[elevmike]

I believe when coils drop out is when the spikes occur. The collapsing field creates the spike.

Rod (the CNC dude)

Yes, That is correct. Just as in the case of an automotive igition coil.

 

[NCPLC]

Power Quality!

More than likely your problem is power quality. This is especially true if the "spikes" are noticed on start-up and e-stop. Get a power analyzer and recorder (Fluke is good) and check those pesky harmonics. A temp short is very unlikely, the pattern would definately be more random if so. But log for a few days and see what happens. A larger ground wire may help but will NOT solve the problem, just diminish it. If this is costing a lot of money, call in some power quality analysts to do the job for you!

 

[guest]

(Quote
Originally posted by Rod
I believe when coils drop out is when the spikes occur. The collapsing field creates the spike.

Rod (the CNC dude)


Yes, That is correct. Just as in the case of an automotive igition coil.


MOV-s or the diodes and let us know what the result will be.

 

[mrdmrd]

Hi Guys

Just to shed a little light from an lab I did in college. If you put about 6 volts on a DC relay if measured with a scope when you release the coil the spike will be enough to fry some things. At one point we measured a spike of around 90 volts.
Inductive loads resist a change in current
Capacitive loads resist a change in voltage
Just a little gee wiz information

 

[kash75]

Spikes on Thermocouple signlas

Need a little help:

I am getting very big spikes on my thermocouple input signals to the PLC( 2-3 degrees F). The thermocouples are in the same conduit as the heater contactors. The contactor voltage is 240 vac. The thermocuples are not shielded. I realize that the proper practice it to have shileds and seperate it from high voltage. Well, here it is not the case because of the old wirings. I have put 25 volt AC capacitors around the input starting from 220 mf to 2000 mf and still getting spikes. At this point I do not know what else to do except rewire everything and seperate it from the contactors. Any suggestions before doing major rewiring? What else could influence the spikes? Should I try bigger capacitors?

Thanks

 

[Terry Woods]

kash75,

You should have started a new thread with this issue.

Anyway... it sounds like your situation includes everything that you are not supposed to do.

Even if you are able to find the right filter arrangement, as soon as you have it, it will begin to degrade, albeit slowly. However, with time and changing conditions, sooner or later, the problem will reoccur. All it takes is for conditions to change... and they will.

You would do best to bite-the-big-one (bullet) and re-route the thermocouple wires in the accepted manner. That will, at least, provide a standard and reasonable basic environment for those signals.

That effort, in itself, will eliminate a large portion of the problems. Then, if necessary, you can apply minor filtering.

There are Dip-Switch controlled (selectable) capacitors out there. You might consider one of those for the minor filtering rather than fooling around with a handful of discrete capacitors.

 

[Bitmore]

I'll try for the sake of being. Voltage spikes on a ground wire are the result of one thing. High RF/EMF energy in close proximity of the grounding conductor (or in a container like a control panel). The reactance of the grounding wire allows it to exibit a resistance, thereby causing a voltage to be realized. It is said to "ring". To eliminate this phenominom, a grounding braid may be in order. Depending on your source of "leaking" energy you may have to install several braid blocks.
The most simple common cause of induced energy is caused by long bonding conductors,or ground conductors not properly shielded from RF/EMF forces, in this case replacing the ground bonding conductor by the braided wire will reduce reactance. In more severe cases where the energy is introduced homogeneously, a wire braid is used throughout the panel. Simple malliable Iron rings may also help, (as I am sure you have seen on Computer power cables) they are easily applied and Radio Shack has several "clam shell" designs to choose from, appy them just before the ground connection to the device suseptable.
Also, solid grounding conductors are the worse wire to use in grounding! Illiminate any solid grounding conductors you may find and replace them with either stranded or braided wire.

Hope this helps,
bitmore

 

[Tom Jenkins]

Kash, at some point it is cheaper to bite the bullet and go back and do things right. You can band-aid till hell freezes over, and when you total up the cost, time, aggravation, and errors it ends up costing more in the end.

Run a new conduit, get the right wire, and take the T/C leads out of that power conduit. I doubt you will be happy with the results otherwise.

"Why is there never enough time to do it right, but always enough time to do it tiwce?"
Anonymous

 

[Pierre]

"Why is there never enough time to do it right, but always enough time to do it tiwce?"Anonymous

Bingo!