Resistance Welder tripping

[GaTech04]

I just returned from a service call at a body shop where they are having problems with breakers tripping when thier resistance welder is in use. Im not too familiar with welders, and after reading through 630.31 I want to make sure Im not overlooking anything.

Here is the nameplate data on the welder:

Duty Cycle Capacity 50% - 24kVa
Voltage - 208V, 50/60Hz, 3 phase
Max Output - 6200A DC
Max Input - 74 kVa
Duty Cycle - 5.6%

The welder has a 50A cord/plug and there are 5 locations with 14-50R outlets available for use. 3 of the 5 outlets are on thier own breakers, the other two share one breaker. All the breakers are 50/3P and the supply conductors are all 8 awg.

Using 630.31A(2) using a value of 7.5% I get 55.5A and 5% is 45A - I assume I could extrapolate an exact current, but 50A seems about right.

The customer says the breaker trips after about 4 welds or so. I'm not sure the duration of the welds, or how that correlates to the duty cycle - but does this look like it would be fixed by increasing the OCP size? If so, is there a rule of thumb for sizing (50 -> 150)?


PART II

The customer also noted that a couple 20/1P breakers trip along with the welder on occasion that are located in a completely different panel. After opening the panel with the welder breakers in it, things became a bit odd.

The panel is a 200A MCP, that is fed with #1 AWG conductors that go to a tap can right below the panel and tap to some 500MCM that go to a 400A/3P breaker in the main switchboard. Both the ground and neutral land on a isolated bar in the panel, with both neutral and ground branch circuit wires landed on it.

Any idea why the breakers in the other panel would be tripping? If I increased the breaker size on the welding circuits (to prevent future trips), would I be leaving or strengthening a hazard?

I hope this makes sense, and any help is appreciated. Please let me know if more information is needed.

 

[LarryFine]

Max Input - 74 kVa

This seems to me to be a concern. I think the 50a breakers are too small.

 

[iwire]

I would start with the manufacturer, either see if you can download the manual for the welder or call on the phone and see if they recommend a certain size breaker for this unit.

 

[GaTech04]

I tried looking up the manufacturers website for a spec sheet, but it wasnt too informative.

http://www.usaspotwelding.com/products_spotwelders.html

Hirane AS-200 F/N

I'll give them a call tomorrow and see if I can get some help.

Thanks guys!

 

[mikeames]

As far as other breakers tripping it may be due to a voltage sag when the welder is running. What type of load do the other 20 amp breakers feed?

 

[GaTech04]

One of the mechanics said that it also tripped 1 or 2 lighting circuits for the storage/office area off the main workshop. He didnt seem completely certain, and I didnt try to replicate the problem.

 

[hillbilly1]

If it is a spot welder, several things come into play with those, voltage drop is one. These type welders draw an large current for a short period of time. contact the manufacture for the proper overcurrent protection. Do the lights dim greatly as a weld is made? This may be causing enough voltage drop to increase the current on other loads. The weld time could be set too long. I had one that was 480 volt 3 phase that was blowing 200 amp fuses occasionally. Watched the process, and the weld time was too long, current should have been increased instead.

 

[GaTech04]

Well I havent had any response from the supplier yet. Would it be appropriate to use the 74kVa nameplate max. input for the rated primary current which infers the following? :

74,000 / (208x1.73) = 205.6A -> 225A min OCP

205.6A x 0.5 = 102.3 A for the supply conductors min. #2 or #1 depending on circuit length

The 225A breaker would force me to upsize the 200A MCB panel that is existing with a 400A MCB panel and tap off the existing 400A feeder that terminates in the JBox below the panel with a #8 or larger wire.

Does this sound about right?

Thanks

 

[iMuse97]

Well I havent had any response from the supplier yet. Would it be appropriate to use the 74kVa nameplate max. input for the rated primary current which infers the following? :

74,000 / (208x1.73) = 205.6A -> 225A min OCP

205.6A x 0.5 = 102.3 A for the supply conductors min. #2 or #1 depending on circuit length

The 225A breaker would force me to upsize the 200A MCB panel that is existing with a 400A MCB panel and tap off the existing 400A feeder that terminates in the JBox below the panel with a #8 or larger wire.

Does this sound about right?

Thanks

Without knowing any other info. (ie. manufacturer's data or specs.), I'd say you've about got it. At the very least you'll keep the welder supplied w/ appropriate power. And I think you will kick the other problems, too.

 

[GaTech04]

What should I do about means of disconnect at the welders?

They currently have N14-50A cord/plug/outlet assemblies that were factory installed. Going by 630.33 I would need a disconnecting means rated at the 102A supply conductor rating.

This also reminds me that there was a built in 50A breaker on the welder - is this normal?

I suppose I need to try harder for some manufacturers specs.

 

[ELA]

Well I havent had any response from the supplier yet. Would it be appropriate to use the 74kVa nameplate max. input for the rated primary current which infers the following? :

74,000 / (208x1.73) = 205.6A -> 225A min OCP

205.6A x 0.5 = 102.3 A for the supply conductors min. #2 or #1 depending on circuit length

The 225A breaker would force me to upsize the 200A MCB panel that is existing with a 400A MCB panel and tap off the existing 400A feeder that terminates in the JBox below the panel with a #8 or larger wire.

Does this sound about right?

Thanks

This is not my normal realm so excuse me if I am off the mark. I was most interested in why your breakers in another panel were tripping? Do you feel these changes would correct that?

From what I have read (have no direct experience here) it seemed as though you should use the 50% nameplate KVA for your calculations unless you know the actual duty cycles they are using.

If you use less than than the 74KVA rating I would also suggest you look into the magnetic trip point of the breakers you use to assure they have a longer time constant to allow for the peak currents of the welder.


I was a little confused by your last line where you said you would tap in the JBox below the panel with a #8 or larger wire.
Didn't you say that the existing tap was a #1awg?


As far as why the other breakers are tripping I would consider taking some voltage and current measurements while the welder is running at both panels.

 

[GaTech04]

We would probably reuse the existing #1 taps to the new panel as long as they reach - if we get a bottom fed 400A MCB they should do fine.

For the nameplate kVa to use, 630.32 says to go by the rated primary current which is defined as the kVA given on the nameplate. Im not sure if you are supposed to use the max input kVA value, the 50% duty value or other though.

Also, 630.32 only calls for the OCP to not exceed 300% of the rating, it doesnt tell you the minimum (Id assume it would be 100% though).

Im really not too familiar with duty cycles, but if we can use the 50% duty cycle kVA rating that would be:

24kVA/(208x1.73) = 66.7A -> 70 to 200A OCP and a 33.3A+ ampacity supply conductor.

If this is the case, then it may be a simple case of replacing the existing breakers with 70/3.

 

[GaTech04]

I think I made a mistake:

66.7A x 0.5 = 33.35A for the conductor (#8)
The MOCP can not be more than 3 times the conductor rating however (630.32 (B)) so the OCP would be 100A.


630.33 would allow me to use the 14-50R as a disconnecting means in this case since it exceeds to #8 ampacity. This section also states I would have to either add a 5th breaker or delete one of the existing outlets so each circuit only supplies 1 welder max.

 

[Brian Oshman]

Duty cycle has to do with the amount of use time in relation to the amount of break time.

In the case of this welder it has a 5.6% duty cycle rating which means for the amount of time it is used which is 5.6% then it will have to remain idle or off for 94.4% longer to prevent damage to the coil from overheating/overloading.

The plug being factory installed and it having a 50 amp breaker on it determines that that would also be your max circuit rating.

Measure the current draw while in use to determine if the circuit is overloaded which may determine if the breaker is bad.

Is this happening on all of the circuits it is used on or just one? Look into installing high inrush breakers of the same size as spot welders do create high inrush for short periods of time. This will prevent nuisance tripping under normal duty cycles. Increasing circuit size is not the answer. Just because it worked for a period of time and then started exhibiting problems may just be due to the wrong breaker application and the high inrush load is wearing out the magnetic portion of the breaker and making it more sensitive.

Check what is connected to the other 20 amp circuits that are tripping. If possible, try those breakers on another phase. Check for loose connections and poor or no grounds.

 

[GaTech04]

Great - thank you for that information.

The mechanic was saying that it would trip after 4 or so welds, of course I have no idea on the duration of the welds, but perhaps he was just exceeding the duty cycle time. He did say that it tripped at several of the locations, but if it's the CB in the panel tripping (which is sized the same as the welder ocp) perhaps the high-inrush would help that.

As far as the other circuits that are tripping, it very well could be a myriad of things. The panel I was mainly focused on wasnt bonded, and the shared ground/neutral bar was isolated from the panel. I did open another panel near the main switchgear and it had a total of 3 ground wires in it (was a full 42 ckt panelboard) which leads me to believe they were using the conduits as the grounding path at the time of install (not sure how old the building is).