Panels breakers tripped...twice

bbbuddy

3 weeks ago we installed 6 more panels, into a separate combiner box and a new run to the controller...

The new panels are Suntech 200w 26.2 Vmp, 33.4 Voc, 7.63 Imp, 8.12 Isc. They are wired 3 in series, each series into a 20 amp Midnight solar breaker, then the new array has it's own wire run to the controller panel.

Other than discovering they were mismarked at the factory (red on the negative MC4 side) they all read ok on our multimeter, and work fine so far...at least until yesterday. We found one series had the breaker tripped. A little checking found a slightly loose connection where #10 wire had been crimped onto an MC4 pigtail.

All connections were checked again after that. This morning BOTH series breakers were tripped...I reset them and the Trimetric and Midnight Classic show the right amount of power in.

Other than somehow getting two more loose connections, which we just checked yesterday....What am I missing?

Vic

Re: Panels breakers tripped...twice

Hi bbb..

Number of things here, but, first, it appears that you have a single CC.

What are the Vmp, Imp of the original PV modules?

A loose connection should not cause a PV Combiner breaker to trip. There is no slightly loose connection. They must all be fery tight, and the best way to ensure this is to use the exact correct crimper. Another concern is the previous mismarking of the PV leads. Normally, these have no polarity marking other on the CORRECT MC-4 connector, in which case, it is molded into the connector body (in my experience). Also, think that the 20 A breaker is a bit large, unless it is speced on the label on the back of the PV module. Would normally expect a 15 A breaker, BUT, this should not affect any of what you are seeing now.

So you are saying that a meter was used to verify that each string of the new PVs shows the correct polarity?

And, does each combiner output have its own circuit breaker at the input of the Charge Controller?

Later, Vic

BB.

Re: Panels breakers tripped...twice

A loose connection on the breaker itself can cause overheating at the breaker (and cause a thermal breaker to trip).

Also old/failing/hot breakers can trip--But this sounds different. Is the breaker box shaded from full sun? No water/rust inside the box?

Is there an issue with grounding (somebody grounded the array negative terminal to the local frame/safety ground)?

-Bill

bbbuddy

Re: Panels breakers tripped...twice

Vic wrote: »

Hi bbb..

Number of things here, but, first, it appears that you have a single CC.

What are the Vmp, Imp of the original PV modules?

A loose connection should not cause a PV Combiner breaker to trip. There is no slightly loose connection. They must all be fery tight, and the best way to ensure this is to use the exact correct crimper. Another concern is the previous mismarking of the PV leads. Normally, these have no polarity marking other on the CORRECT MC-4 connector, in which case, it is molded into the connector body (in my experience). Also, think that the 20 A breaker is a bit large, unless it is speced on the label on the back of the PV module. Would normally expect a 15 A breaker, BUT, this should not affect any of what you are seeing now.

So you are saying that a meter was used to verify that each string of the new PVs shows the correct polarity?

And, does each combiner output have its own circuit breaker at the input of the Charge Controller?

Later, Vic

Original pv are 8 100w 75 Vmp, 1.34 Imp modules, wired in parallel to a combiner box. New panels are in a new combiner box, new breakers. Both have their own run to a MNDC panel, each with it's own breaker before combining into the Classic 150 cc.

The panels had factory installed red tape wrapped around a lead, and was on the negative side, but we discovered that and notified the seller (sunelec.com) so they could warn others. Yes, we used a multimeter for polarity...

No shading anywhere near. No negative ground. This array worked fine for 3 weeks prior.

The breaker size is because I read on here that the size should be about 2 times Imp of one panel, and then add in 25%, go to next higher size.

2x7.63=15.26 + 25% = 19.07 = 20 amp.

Each array wire run is about 90 feet at 75 volts, 13 and 15 amps respectively, and is #2 aluminum house entrance wire. NOALOX used at al/cu connections.

Cariboocoot

Re: Panels breakers tripped...twice

So the original array is 75 Vmp @ 10.72 Imp and it is paralleled to the new array which is 78.6 Vmp @ 15.26 Imp?

And somehow this is causing 20 Amp breakers to trip?

Does this set up have the DC GFI system?

Vic

Re: Panels breakers tripped...twice

Another question, Are the PVs all installed at the same elevation angle, and same direction -- are they receiving the same sun at the same time? Know that you mentioned, "no shading" ...

And what is the exact part number of the new PV modules?

BB, certainly loose breaker connections can cause heating and breaker trips. I was reading the OP as loose MC-4 connector/s.

I had felt that the PV breaker was 1.56 X Isc, or up to the Max spec on the PV label. Vic

bbbuddy

Re: Panels breakers tripped...twice

Cariboocoot wrote: »

So the original array is 75 Vmp @ 10.72 Imp and it is paralleled to the new array which is 78.6 Vmp @ 15.26 Imp?

And somehow this is causing 20 Amp breakers to trip?

Does this set up have the DC GFI system?

All paralleled panels or series are into breakers in combiner boxes.

The older array is 8 paralleled panels into a combiner box. The new array is two series strings of 3, into a new combiner box.

Coot, Each array is run on it's own wire from two combiner boxes and into it's own breaker (two 20 amp Breakers) at the MNDC box before CC. They are then combined by a "jumper" at the hot output of the two breakers and into CC on a short #4 wire. No external DC GFI.

Negatives are combined by a split bolt into a #4 wire to the negative side of the 500 amp shunt in the MNDC.

bbbuddy

Re: Panels breakers tripped...twice

Vic wrote: »

Another question, Are the PVs all installed at the same elevation angle, and same direction -- are they receiving the same sun at the same time? Know that you mentioned, "no shading" ...

And what is the exact part number of the new PV modules?

BB, certainly loose breaker connections can cause heating and breaker trips. I was reading the OP as loose MC-4 connector/s.

I had felt that the PV breaker was 1.56 X Isc, or up to the Max spec on the PV label. Vic

Same angle, same direction, no shading, except that one of the pole mount 100 watt set of mono's (4 panels) shades the other set of 4 panels until the sun is up about 8 am. But this array has been fine since we put it in last year, and the breakers on this array have never tripped.

It's the new array that has both breakers tripped this morning. The panels are all set for fall, and will be reset for winter next week as per http://www.macslab.com/optsolar.html

The single "loose" wire (meaning a yank on it pulled it loose)we found was a not-so-good-crimp at a MC4 pigtail to #10 wire. The connections at the breakers are all very tight.

ggunn

Re: Panels breakers tripped...twice

bbbuddy wrote: »

The breaker size is because I read on here that the size should be about 2 times Imp of one panel, and then add in 25%, go to next higher size.

The correct way to do it is to multiply Isc times 1.56 (1.25 twice) and go to the next fuse/breaker size up. 1.56 X 8.12 = 12.7, next size up is 15A. But with only two strings, you don't need string OCPD's.

i have to insert here that these 2 new strings are being paralleled to a bunch of high voltage pvs that are already paralleled. each of those hi v pvs needs individually fused as well as each string from the new pvs. niel

Vic

Re: Panels breakers tripped...twice

bbb..

Have never seen a 100 W high-ish voltage (75V) Mono PV, what is the model number for these PVs?

What is the model number for the New Suntech PVs?

Have you opened the circuit breakers on each string of the new PVs, and measured the Open Circuit Voltage at the input side of the breakers? Does this agree with the nominal Voc of each PV X 3? The polarity of them must be correct, or the breaaker would never be closed with the sun up, and the Trimetric would never read correctly.

Interesting! Thanks, Vic

Cariboocoot

Re: Panels breakers tripped...twice

ggunn wrote: »

The correct way to do it is to multiply Isc times 1.56 (1.25 twice) and go to the next fuse/breaker size up. 1.56 X 8.12 = 12.7, next size up is 15A. But with only two strings, you don't need string OCPD's.

Curiously it is the 20 Amp breakers on the new array that are tripping (if I understand correctly). With their Isc of 8.12 the breakers should be 15 Amp per string according to the NEC regs. In theory it would be impossible for a single string to output that much. The 'old' array is capable of 10+ Amps and the new one capable of 8 Amps per string so there is the slight possibility that enough current could be back-fed to one of these strings to trip the 20 Amp breaker. It would have to be a weird confluence of circumstances though.

Measuring current with a clamp-on DC Ammeter would probably result in endless hours of watching the problem not occur.
Installing blocking diodes per string to prevent any current back-feed would solve the problem if that is the problem.

bbbuddy

Re: Panels breakers tripped...twice

Vic wrote: »

bbb..

Have never seen a 100 W high-ish voltage (75V) Mono PV, what is the model number for these PVs?

What is the model number for the New Suntech PVs?

Have you opened the circuit breakers on each string of the new PVs, and measured the Open Circuit Voltage at the input side of the breakers? Does this agree with the nominal Voc of each PV X 3? The polarity of them must be correct, or the breaaker would never be closed with the sun up, and the Trimetric would never read correctly.

Interesting! Thanks, Vic

The 100 watt panels are DuPont DA 100-A1 thin film amorphous

The Suntech are STP200-18/Ud 200 watt poly

Yes, we measured voltage both with and without the breakers being open. Open they were in the 90 + volt range, closed they were depending on the time in the 60+v range.

I just got done with a one hour equalize after my batteries reached float, while at the same time did two loads of laundry, the bread machine is running, my satellite internet is on, and I had more power available if I wished...(Turning lights on resulted in more watts out at the pv) so with the breakers on everything seems to work as it should.

bbbuddy

Re: Panels breakers tripped...twice

Cariboocoot wrote: »

Curiously it is the 20 Amp breakers on the new array that are tripping (if I understand correctly). With their Isc of 8.12 the breakers should be 15 Amp per string according to the NEC regs. In theory it would be impossible for a single string to output that much. The 'old' array is capable of 10+ Amps and the new one capable of 8 Amps per string so there is the slight possibility that enough current could be back-fed to one of these strings to trip the 20 Amp breaker. It would have to be a weird confluence of circumstances though.

Measuring current with a clamp-on DC Ammeter would probably result in endless hours of watching the problem not occur.
Installing blocking diodes per string to prevent any current back-feed would solve the problem if that is the problem.

Coot, when researching here, I copied and pasted the following from this forum:

"Also, each series string of panels should have a fuse/breaker ~2xIsc of the series panels. This is "required" for arrays with three or more parallel strings to prevent panel/wiring short circuits from starting fires (other panels supply energy into shorted string)." I didn't save who wrote it though.

So that is how I ended up with 20 amp breakers. With breakers at the controller for each array, and breakers at the combiner boxes, how can there be any backfeeding at all?

bbbuddy

Re: Panels breakers tripped...twice

Ah, I see it was BB (Bill) that I copied/pasted the info on breaker size; he also has stated:

"each panel should have an appropriately rated fuse (based on the panel's spec. for Series Rated Fuse or on ~2xIsc rating, or ~2.2xImp)"

which again would be a 20 amp breaker per string. 20 amps is the "Maximum Series Fuse Rating" for the panel, too.


The entire part I copied/pasted from BB (Bill) was :

"Very roughly, for optimum power, when placing panels in series you should match Imp to within 10%. For panels in parallel, you should match Vmp to within 10%... The farther you get from 10%, the more power loss (than 10%) you may get (limit overall production of the array, and can confuse the MPPT controllers with multiple power "power peaks" in the Power vs Current curve).

Also, each series string of panels should have a fuse/breaker ~2xIsc of the series panels. This is "required" for arrays with three or more parallel strings to prevent panel/wiring short circuits from starting fires (other panels supply energy into shorted string)."

from here: http://forum.solar-electric.com/archive/index.php/t-9175.html

Cariboocoot

Re: Panels breakers tripped...twice

bbbuddy wrote: »

Coot, when researching here, I copied and pasted the following from this forum:

"Also, each series string of panels should have a fuse/breaker ~2xIsc of the series panels. This is "required" for arrays with three or more parallel strings to prevent panel/wiring short circuits from starting fires (other panels supply energy into shorted string)." I didn't save who wrote it though.

I don't know who wrote it either. The NEC reg is like ggunn said: Isc * 1.25 * 1.25
Since a panel (or string of panels) can not produce more than its Isc, this protection applies only to three or more parallel connections where it is possible for two (or more) of the panels/strings to produce 2X the third's maximum should that third one become shorted. This is based on equal capacity panels/strings. When they are unequal in current you need to provide protection for the panel/string based on its Isc so that should a problem arise additional panels can't force more current than the max it can handle.

So that is how I ended up with 20 amp breakers. With breakers at the controller for each array, and breakers at the combiner boxes, how can there be any backfeeding at all?

You have separate combiners & wires, but in the end both arrays connect to the same charge controller, correct? Thus in fact all the panel strings are in parallel; the final common connection point being the controller's input terminals.

Here's where the problem arises: the 'old' panels have a very high Voltage rating compared to the 'new' panels. It is just possible that the new panels can't take the higher Voltage even though when in series they have a collective higher Voltage. Panels are diodes, as it were, and if you apply enough Voltage a diode will 'break down' and allow current to flow backwards. Normally this isn't a problem for most arrays because the panels are all matched for Voltage, current, and shunt resistance. But it may crop up on mismatched strings under some circumstances.

I'm still not saying this is absolutely the problem; just a possibility, and a remote one at that.

Cariboocoot

Re: Panels breakers tripped...twice

BTW, I don't usually disagree with Bill but as far as off-grid panel fusing is concerned the NEC regs and/or 2X Isc is really redundant since the panels can't put out more than Isc. If you have three in parallel and one even partially shorts having it fused for Isc (size up to nearest available) prevents the other two from feeding it 2X its maximum current. Under normal circumstances these 'too small' fuses will not present any trouble at all.

The best info is provided by the manufacturer who should give a spec on "series fuse rating". Hard to beat their recommendations.

bbbuddy

Re: Panels breakers tripped...twice

Cariboocoot wrote: »

I don't know who wrote it either. The NEC reg is like ggunn said: Isc * 1.25 * 1.25
Since a panel (or string of panels) can not produce more than its Isc, this protection applies only to three or more parallel connections where it is possible for two (or more) of the panels/strings to produce 2X the third's maximum should that third one become shorted. This is based on equal capacity panels/strings. When they are unequal in current you need to provide protection for the panel/string based on its Isc so that should a problem arise additional panels can't force more current than the max it can handle.



You have separate combiners & wires, but in the end both arrays connect to the same charge controller, correct? Thus in fact all the panel strings are in parallel; the final common connection point being the controller's input terminals.

Here's where the problem arises: the 'old' panels have a very high Voltage rating compared to the 'new' panels. It is just possible that the new panels can't take the higher Voltage even though when in series they have a collective higher Voltage. Panels are diodes, as it were, and if you apply enough Voltage a diode will 'break down' and allow current to flow backwards. Normally this isn't a problem for most arrays because the panels are all matched for Voltage, current, and shunt resistance. But it may crop up on mismatched strings under some circumstances.

I'm still not saying this is absolutely the problem; just a possibility, and a remote one at that.

Coot, before I ever bought the new panels I ran the proposed addition by the forum and you said "I don't see any flaw in the plan, then."
http://forum.solar-electric.com/showthread.php?16537-Expanding-system-panel-mix-ok

And if the 75 volts from the 100 watt panels could hurt these new panels, so could the higher 78.6 voltage from one string hurt the panels in the other string, yet panels are series-ed and paralleled all the time. This just doesn't make sense to me...

BB.

Re: Panels breakers tripped...twice

The ~2x Isc rating and finding the closest fuse/breaker was for panels where they never listed the Series Fuse requirement (typically non-UL/NRTL listed panels).

The Dupont thin film panels are a special case... They only output 1-2 amps or so at 70 volts and have a series protection fuse on the order of 15 amps. So you can place a lot more than three panels in parallel without a series protection fuse.

Most other panels have the Series Protection fuse just under 2*Isc, and why we have to use a series protection fuse for 3 or more panels in series (i.e., two panels * Isc exceeds the fuse rating into a third shorted panel).

-Bill

bbbuddy

Re: Panels breakers tripped...twice

BB. wrote: »

The ~2x Isc rating and finding the closest fuse/breaker was for panels where they never listed the Series Fuse requirement (typically non-UL/NRTL listed panels).

The Dupont thin film panels are a special case... They only output 1-2 amps or so at 70 volts and have a series protection fuse on the order of 15 amps. So you can place a lot more than three panels in parallel without a series protection fuse.

Most other panels have the Series Protection fuse just under 2*Isc, and why we have to use a series protection fuse for 3 or more panels in series (i.e., two panels * Isc exceeds the fuse rating into a third shorted panel).

-Bill

The DuPont panels did not have a Maximum Series Fuse Rating on them, but I called DuPont and was told it was 8 amps.

These new panels have a Maximum Series Fuse Rating of 20 amps.

Cariboocoot

Re: Panels breakers tripped...twice

bbbuddy wrote: »

Coot, before I ever bought the new panels I ran the proposed addition by the forum and you said "I don't see any flaw in the plan, then."
http://forum.solar-electric.com/showthread.php?16537-Expanding-system-panel-mix-ok

And if the 75 volts from the 100 watt panels could hurt these new panels, so could the higher 78.6 voltage from one string hurt the panels in the other string, yet panels are series-ed and paralleled all the time. This just doesn't make sense to me...

No, it should not be a problem. As I said before this is not necessarily what the problem is. And if it is, it is due to some odd circumstances (new panel(s) shaded, old panels in full sun sort of thing). It shouldn't hurt the panel. Darn nuisance if it keep happening of course.

Vic

Re: Panels breakers tripped...twice

Well, everyone is correct on the fuse rating:
http://www.arcmansolar.com/Assets/Products/Documents/STP200_18Ud.pdf

The mfg states "Maximum Fuse Rating 20A"

I personally like 1.56 X Isc, as that is safe, and might stress the PV a bit less in case of a fault.

Thanks bbb for the added info on the PVs. Seems impossible for the DuPont PV to create enough current to trip the 20 A breaker.

Out of curiosity, what brand are the combiner breakers for the new PVs?

Wonder if the red tape on one PV lead was an indication of another issue, not polarity?? These were supposed to be full spec PVs were'nt they? More interesting, yet. Vic

bbbuddy

Re: Panels breakers tripped...twice

Vic wrote: »

Well, everyone is correct on the fuse rating:
http://www.arcmansolar.com/Assets/Products/Documents/STP200_18Ud.pdf

The mfg states "Maximum Fuse Rating 20A"

I personally like 1.56 X Isc, as that is safe, and might stress the PV a bit less in case of a fault.

Thanks bbb for the added info on the PVs. Seems impossible for the DuPont PV to create enough current to trip the 20 A breaker.

Out of curiosity, what brand are the combiner breakers for the new PVs?

Wonder if the red tape on one PV lead was an indication of another issue, not polarity?? These were supposed to be full spec PVs were'nt they? More interesting, yet. Vic

The new combiner breakers are Midnight Solar. Sunelec.com agreed with me that the red tape was SUPPOSED to indicate the positive side, and they were going to notify other customers and the factory. Yes, they are full UL listed grade a panels, NOT "grade b" panels.

Vic

Re: Panels breakers tripped...twice

Hi bbb..

OK, sorry for all the questions.

It does appear that you have tried very hard to do things correctly, even asking questions here before buying hardware, and supporting the Forum Host.

Earlier, someone wondered about a ground fault of some kind. I dunno about your situation.

If you could, would you confirm or re-confirm the label on these new PV breakers, to make certain that they are what you ordered. Just to make sure. Good Luck, Vic

bbbuddy

Re: Panels breakers tripped...twice

Well, I can go out and open up the box, but just to let you know, I bought them from NAWS...the MNEPV series


Yep, just read MNEPV right on the breakers....those are Midnight Solar breakers

Don't worry about asking me questions, I want to get to the bottom of this...

Vic

Re: Panels breakers tripped...twice

bbb..

Sorry. I did not state the request clearly.

Was wondering if the breakers that were shipped were the wrong current rating. This is not likely. Some of the MN breakers use some tape on the face, which has a number that corresponds to the Amp rating of the breaker. Some have it painted on the handle. I forget, was just wondering if you could confirm the Amp rating of the breakers that are in the new strings of PVs.

All of the new MNPV breakers I have are at another site. Dunno, it is curious. Thanks, Good Luck, Vic

Cariboocoot

Re: Panels breakers tripped...twice

Intermittent problems are the most difficult to find. If the breaker would trip every time it got flipped on it would be a piece of cake to analyze!
This could be due to edge-of-cloud events where the cloud is blocking light on one string and concentrating it on the other(s).

BB.

Re: Panels breakers tripped...twice

Ye 'ol DC Current Clamp Meter (Sears $60) will tell you if you are running close or not.

I know that the bad connection was a MC-4 Crimp--My suggestion was to look for poor/overheating connections at the breakers themselves (which can cause failures).

-Bill

bbbuddy

Re: Panels breakers tripped...twice

Vic wrote: »

bbb..

Sorry. I did not state the request clearly.

Was wondering if the breakers that were shipped were the wrong current rating. This is not likely. Some of the MN breakers use some tape on the face, which has a number that corresponds to the Amp rating of the breaker. Some have it painted on the handle. I forget, was just wondering if you could confirm the Amp rating of the breakers that are in the new strings of PVs.

All of the new MNPV breakers I have are at another site. Dunno, it is curious. Thanks, Good Luck, Vic

The breakers are labeled 20 amps right on the on/off handles

bbbuddy

Re: Panels breakers tripped...twice

BB. wrote: »

Ye 'ol DC Current Clamp Meter (Sears $60) will tell you if you are running close or not.

I know that the bad connection was a MC-4 Crimp--My suggestion was to look for poor/overheating connections at the breakers themselves (which can cause failures).

-Bill

We have a Craftsman multimeter, it measures both DC and AC voltage, amps, ohms, etc etc, and is also a clamp meter....AC and DC

bbbuddy

Re: Panels breakers tripped...twice

Today all seems well with the panels. So far anyway.

Vic

Re: Panels breakers tripped...twice

Hi bbb...

Thanks for checking the rating marked on the breakers.

Well, at this point, I am back in the "loose connection causing breaker to overheat/trip" camp. Just seems that there is NO source of enough current to cause the breaker to trip on the input side of the CC. All of the CBI DIN-Rail breakers have hidden screws that drive the clamp that connects the wires/bussbars. These screws are small, and the breakers are not particulary solidly mounted mechanically. So, one needs to FIRMLY grip the breaker when torquing these screws. AND, these connections of each breaker terminal need to be RE-TOQUED after a period of time. These hidden screws will accept a larger sized Phillips driver bit than would be imagine when looking at the hole in the breaker for these screws (IMHO).

Others have noted this. It would be a good idea to check/re-torque these screws as aggressively as you possibly can. And of course, this should be done when there is NO PV VOLTAGE PRESENT. Best Of Luck, Vic