logo elektroda
logo elektroda
X
logo elektroda
Advanced Search
logo elektroda

PV Installation Grounding on Steel Roof Tiles: Contractor Advice, TN-S System & Regulations

ozzi132 23253 46
Best answers

Should PV module frames and mounting rails on a steel-tile roof in a TN-S building be additionally bonded to PE, and is there any legal requirement to do so?

The contractor’s “no extra bonding” position is not the safest answer: the module frames, supporting structure, and metal roof covering should be included in equipotential bonding to the PE/GSU [#18644348][#18644720][#18645225] Forum users cite PN-HD 60364-7-712, PN-HD 60364-5-534, PN-EN 62305 and related SPD standards as the relevant technical basis, and one recommendation was to use a 16 mm² Cu conductor and grounding lugs to connect the PV structure to earth/PE [#18644340][#18644456] At the same time, one poster reports that UDT said there is no separate legal obligation “as such,” so the issue is mainly a standards-and-safety matter rather than a simple statutory yes/no [#18643634] If the building already has a foundation earth electrode, don’t create an extra rod just for the PV; bond the PV parts to the existing equipotential system instead [#18644720] The thread also recommends surge protection on the DC side, placed as close as possible to the array/inverter, especially if lightning protection is present or considered [#18645225]
Generated by the language model.
ADVERTISEMENT
Treść została przetłumaczona polish » english Zobacz oryginalną wersję tematu
Report a violation of the law
Reply | New topic
  • #1 18643055
    ozzi132
    Level 10  
    Hello,

    I have made a PV installation on a roof covered with steel tiles. The building has installations in the TN-S system and does not have a lightning protection system.
    The inverter is grounded to the PE bus. As for the panels and structures themselves, the contractor claims that there is no need for additional installation on the roof.

    Is the contractor right in this case? Are there any laws regulating this?
  • ADVERTISEMENT
  • #3 18643145
    jdubowski
    Tube devices specialist  
    ozzi132 wrote:
    The inverter is grounded to the PE bus.


    And the PE rail grounded to what?

    ozzi132 wrote:
    As for the panels and structures themselves, the contractor claims that there is no need for additional installation on the roof.


    The supporting structure and the metal roof covering must also be covered with a bonding connection.

    andrzej20001 wrote:
    What will happen when the panel gets a puncture on the frame, roof and goes a few hundred volts. ???


    Exactly...
  • #4 18643149
    ozzi132
    Level 10  
    The contractor says they don't have to. As for the puncture, the contractor claims that nothing will happen ... because it's direct current ...

    Added after 3 [minutes]:

    jdubowski wrote:
    ozzi132 wrote:
    The inverter is grounded to the PE bus.


    And the PE rail grounded to what?



    To the hoop in the foundation of the building.
  • #5 18643172
    andrzej20001
    Level 43  
    Dude, direct current is just a massacre, change your spies better. Gosh
  • #6 18643205
    ozzi132
    Level 10  
    In addition, I attach a diagram of how it looks like.

    PV Installation Grounding on Steel Roof Tiles: Contractor Advice, TN-S System & Regulations
  • #7 18643522
    andrzej20001
    Level 43  
    A meter of cable 5-6 PLN plus a pylon, a bit of work with thermotube eyelets, etc. plus screws, inox nuts, let it be PLN 300 cuzamen to buy.
    Maybe to be sure call the UDT.
  • #8 18643634
    ozzi132
    Level 10  
    I called, from the information I received there is no legal requirement as such ...
  • #9 18643813
    marekj31
    Level 23  
    Quote:
    What will happen when the panel gets a puncture on the frame, roof and goes a few hundred volts. ???

    Well, what can happen when it is grounded and goes a few hundred volts constant ??
  • #10 18643991
    andrzej20001
    Level 43  
    That's when you grab the gutters at the bottom
  • ADVERTISEMENT
  • #11 18644329
    ozzi132
    Level 10  
    So what do you advise?
  • #12 18644340
    noja102
    Level 24  
    marekj31 wrote:
    Quote:
    What will happen when the panel gets a puncture on the frame, roof and goes a few hundred volts. ???

    Well, what can happen when it is grounded and goes a few hundred volts constant ??
    You do not swing your legs as with a variable one, and the standards are PN-EN 62305-1, PN-EN 62305-2, PN-EN 62305-4, PN-EN 61643-11, PKN-CLC / TS 61643-12, PN-EN 61643-21, PN-EN 61643-22, CLC / TS 50539-12, PN-HD 60364-7-712, PN-HD 60364-5-534. And ask this contractor if he will give you a paper declaration and will guarantee it with his private property
  • ADVERTISEMENT
  • #13 18644348
    3301
    Level 34  
    ozzi132 wrote:
    So what do you advise?


    Equipotential bonding: PE rail - module frames
  • ADVERTISEMENT
  • #15 18644720
    3301
    Level 34  
    If there is a foundation earth electrode, why double and make an additional one with a pin, we are talking about an equipotential bonding for protection against electric shock in the event of a dangerous voltage appearing on the photovoltaic module housing
  • #16 18644739
    ozzi132
    Level 10  
    3301 wrote:
    If there is a foundation earth electrode, why double and make an additional one with a pin, we are talking about an equipotential bonding for protection against electric shock in the event of a dangerous voltage appearing on the photovoltaic module housing

    I also think that there is no such need. I have a junction box outside the building. The only problem I have is to get the installer to make this connection ...

    Below is a quote from the email I received from them ...

    In response to the received earthing message, we would like to inform you that in accordance with applicable regulations, there is no obligation to ground panels and structures under the panels. Roof solar structures are not accessible conductive parts and there is no obligation to cover them with equipotential bonding. The legislator imposes the necessity to ground such structures only if there is a lightning protection device (lightning protection system) on the building.

    When it comes to grounding, we make them in electrical installations that do not have grounding (made in the TN, TN-C system). In your case, the photovoltaic circuit was connected to the existing power circuit in the building, made in the TN-S network system, i.e. with a separate PE conductor (grounding). Its effectiveness was measured (grounding resistance measurement - photo in the appendix) and it was found that it had the correct parameters. Therefore, it was not necessary to make another earthing or improve the existing earthing.
  • #17 18645225
    Jarek Kordalewski
    Level 18  
    3301 wrote:
    If there is a foundation earth electrode, why double and make an additional one with a pin, we are talking about an equipotential bonding for protection against electric shock in the event of a dangerous voltage appearing on the photovoltaic module housing


    And how would this voltage appear (in relation to the ground) if the panel is insulated from the ground without equipotential bonding? Let's not confuse the two. Protection against direct voltage shock must be provided in a different way (primarily insulation and equalization of the potentials of the conductive surfaces remaining within the touch range) and the earthing of the panels de facto increases the risk of a direct voltage shock (because this is when a potential difference of potentials between the ground and DC electrodes may arise) . This risk increases especially if these other protection measures are not taken.

    Of course, this does not mean that the lightning protection of the installation cannot or should not be used. But this protection should primarily rely on direct protection reducing the risk of lightning current flowing directly through the conductive elements of the panels. So in practice it should be a lightning protection system installed above the panels and covering their entire surface with its protective zone. Depending on the concept of protection zones and local conditions, the frame of the panels can (and generally should) be connected to the ground of the object but as an element of protection against indirect lightning effects. That is, as equipotential bonding and discharge of residual currents, not the main discharge current. Of course, in such a case all the more you have to use overvoltage limiters between the two DC electrodes and the ground. Such limiters should be installed as close as possible to the panel connection and possibly repeated at the inverter input.
  • #18 18646182
    theo33
    Level 27  
    Jarek Kordalewski wrote:
    3301 wrote:
    If there is a foundation earth electrode, why double and make an additional one with a pin, we are talking about an equipotential bonding for protection against electric shock in the event of a dangerous voltage appearing on the photovoltaic module housing


    And how would this voltage appear (in relation to the ground) if the panel is insulated from the ground without equipotential bonding? Let's not confuse the two. Protection against direct voltage shock must be provided in a different way (first of all, insulation and equalization of the potentials of the conductive surfaces remaining within the touch range) and the earthing of the panels de facto increases the risk of a direct voltage shock (because this is when a potential difference of potentials between the ground and DC electrodes may arise) . This risk increases especially if these other protection measures are not taken.

    Of course, this does not mean that the lightning protection of the installation cannot or should not be used. But this protection should primarily rely on direct protection reducing the risk of lightning current flowing directly through the conductive elements of the panels. So in practice it should be a lightning protection system installed above the panels and covering their entire surface with its protective zone. Depending on the concept of protection zones and local conditions, the frame of the panels can (and generally should) be connected to the ground of the object but as an element of protection against indirect lightning effects. That is, as equipotential bonding and discharge of residual currents, not the main discharge current. Of course, in such a case all the more you have to use overvoltage limiters between the two DC electrodes and the ground. Such limiters should be installed as close as possible to the panel connection and possibly repeated at the inverter input.


    In such a photovoltaic module (panel) and in any electrical device, especially in a conductive housing, the insulating layer separating the cell poles from the housing of such a module may be damaged, and in this case, with modules connected in series in such a string, connected to the input of a modern inverter, i.e. without galvanic isolation, this voltage in relation to the ground will depend on the place of this damage, and can reach even 1000V and in the case of installation on a conductive roof, it is even more necessary to connect to the ground, and it is not related to lightning protection
  • #19 18646421
    hupo
    Level 14  
    Is it not that at the time of startup one of the steps is that the inverter is checking for a DC fault from the panels? If the panel casing is insulated from the ground, the inverter cannot detect such damage. In addition, it is slippery and more uncertain because there are different RCDs and sometimes they do not exist at all, through the inverter without galvanic isolation (so now probably all string), a clear DC breakdown to the ground will activate the RCD fuse.
  • #20 18646560
    noja102
    Level 24  
    https://pl.aliexpress.com/item/4000571197868....id=1007.1b7576&scm_id=1007.157576.0-7v1.14976. -44ee-99b3-6344b4451bf2 & _t = gps-id: pcDetailBottomMoreOtherSeller, scm-url: 1007.14976.157518.0, pvid: 6c5c7bb5-e2fb-44ee-99b3-6344b4451bf8% 2359686% 235968% 235968% 235968% 235968% 235968% 235968 % 233 325% 237_4976% 230% 23157518% 230_4976% 232711% 237538% 23318_668% 232717% 237 565% 23726 also https://allegrolokalnie.pl/oferta/lgy-16mm2 or https://allegro.pl/oferta/pociąg- cable-linka-lgy-h07v-k-16mm2-yellow-green-5163953623 and a few more https://allegro.pl/oferta/koncowka-kablowa-miedziana-16-oczko-fi8-16mm2-m8-8973534867 and we have peace is it really that expensive? Better if he stabs someone or kills someone with 500V constant, it is the cheapest. What about you people, I gave you the first 3 standards that apply to the installer and there is no discussion with them, ask him to the UDT. This is a freshly made electrical installation, whatever its specificity, it has to adapt to the applicable standards and that's it
  • #21 18646591
    Krzysztof Reszka
    VIP Meritorious for electroda.pl  
    noja102 wrote:
    You do not swing your legs as with a variable one, and the standards are PN-EN 62305-1, PN-EN 62305-2, PN-EN 62305-4, PN-EN 61643-11, PKN-CLC / TS 61643-12, PN-EN 61643-21, PN-EN 61643-22, CLC / TS 50539-12, PN-HD 60364-7-712, PN-HD 60364-5-534. And ask this contractor if he will give you a paper declaration and will guarantee it with his private property

    Of course he will write. Have you mentioned so many standards only that are useful for PV installations?
    ozzi132 wrote:
    Roof solar structures are not accessible conductive parts and there is no obligation to cover them with equipotential bonding

    This is not entirely true because the PN-HD 60364-7-712 standard does not mention the obligation to connect the above-mentioned installations with the grounding terminal, but there are other provisions in this standard.
    Especially 712-410-101 saying that the panels on the DC side should always be considered active even though they are not connected to the grid or to the inverter. This has to do with the next point.
    712-513-101 it speaks of safety during panel operation and servicing, e.g. during snow removal or repair. And then in item 712-542-101 we read about equipotential bonding that can be mounted and the reason for their installation could be the reason I gave you. As you can see, this is not included in the project, so you should get along with them on this issue.
    What DC voltage your panels are issuing.
  • #22 18646638
    noja102
    Level 24  
    The following installation diagram shows: "the expected voltage value is max. 500 V DC"

    Added after 5 [minutes]:

    If on a sunny May day the rain falls from a cloud that does not cover this sun and the panels will have a puncture on the metal roof and someone touches the metal gutter while hiding under the eaves, nothing will happen?
  • #23 18646723
    Krzysztof Reszka
    VIP Meritorious for electroda.pl  
    noja102 wrote:
    If on a sunny May day the rain falls from a cloud that does not cover this sun and the panels will have a puncture on the metal roof and someone touches the metal gutter while hiding under the eaves, nothing will happen?

    The author writes nothing about the metal gutter.
    noja102 wrote:
    "the expected voltage value is max. 500 V DC"

    I have seen too, but is it sure? Often, such projects are prefabs.
  • #24 18646810
    noja102
    Level 24  
    17x Voc 40.35 = 685.95 I 17x Vmpp 34.08 = 579.36 so almost 80V over the 500 and just over that can kill. Even if this gutter will be made of plastic, it will be safe ??? I just mean how much more expensive it would be with this earthing of the panels. And then installers are surprised that there are such regulations as https://www.elektroda.pl/rtvforum/topic3650420.html and that you will have to pass the UDT RES exam and it is not enough to buy SEP here https: //www.olx.pl/oferty/q-SEP/

    Added after 12 [minutes]:

    It is more expensive for the installer if he bought these DEVOTIONALS in wholesale to do it in accordance with the art, having a team on site, maybe an hour longer assembly. It is a disastrous saving for me
    Attachments:
    • twinplus_60_mono.pdf (933.78 KB) You must be logged in to download this attachment.
  • #25 18646857
    Krzysztof Reszka
    VIP Meritorious for electroda.pl  
    Look in the design, even the inverter has no grounding. The drawing sucks too, so let's wait.
  • #26 18646880
    noja102
    Level 24  
    There is a statement by the installer to the DSO at Z-MI, according to Article 233 § 6, how did he send the finished product in bulk? testified not the truth?
  • #27 18647357
    Jarek Kordalewski
    Level 18  
    theo33 wrote:
    In such a photovoltaic module (panel) and in any electrical device, especially in a conductive housing, the insulating layer separating the cell poles from the housing of such a module may be damaged, and in this case, with modules connected in series in such a string, connected to the input of a modern inverter, i.e. without galvanic isolation, this voltage in relation to the ground will depend on the place of this damage, and can reach even 1000V and in the case of installation on a conductive roof, it is even more necessary to connect to the ground, and it is not related to lightning protection


    How is there supposed to be a voltage with respect to earth if both DC electrodes are isolated from this earth? So what if the housing will have a voltage of 1000 V (or any amount of V) to the other electrode? How is the current and shock to flow if I only touch the module housing and the ground?

    Added after 7 [minutes]:

    noja102 wrote:
    how on a sunny May day will rain fall from a cloud that does not cover this sun and the panels will have a puncture on the metal roof and someone touches the metal gutter while hiding under the eaves, nothing will happen?


    What is going to happen? After all, the DC voltage is produced between the electrodes of the panel and not with respect to the ground. For the shock to occur, you would have to touch the other (ungrounded) electrode. That is why I wrote above that grounding the panel housing even increases the risk of electric shock (which does not mean that I do not recommend it).
  • #28 18647586
    edziu
    Level 29  
    e.g. one PV arrester is damaged, an inverter, the DC insulation is bitten by rodents, and other mechanical damage to the insulation, etc.
  • #29 18647694
    Jarek Kordalewski
    Level 18  
    edziu wrote:
    e.g. one PV arrester is damaged, an inverter, the DC insulation is bitten by rodents, and other mechanical damage to the insulation, etc.


    Agreed, but it remains in force that it takes at least two failures for an electrocution to occur and that introducing earth potential on naturally insulated components always increases, not reduces the risk. Which does not mean, I will repeat it again that I do not recommend equalizing the potentials of the panel frames. But everything must be done wisely, properly coordinating the security and equipotential bonding.
  • #30 18647861
    theo33
    Level 27  
    Jarek Kordalewski wrote:
    theo33 wrote:
    In such a photovoltaic module (panel) and in any electrical device, especially in a conductive housing, the insulating layer separating the cell poles from the housing of such a module may be damaged, and in this case, with modules connected in series in such a string, connected to the input of a modern inverter, i.e. without galvanic isolation, this voltage in relation to the ground will depend on the place of this damage, and can reach even 1000V and in the case of installation on a conductive roof, it is even more necessary to connect to the ground, and it is not related to lightning protection


    How is there supposed to be a voltage with respect to earth if both DC electrodes are isolated from this earth? So what if the housing will have a voltage of 1000 V (or any amount of V) to the other electrode? How is the current and shock to flow if I only touch the module housing and the ground?

    Added after 7 [minutes]:

    noja102 wrote:
    how on a sunny May day will rain fall from a cloud that does not cover this sun and the panels will have a puncture on the metal roof and someone touches the metal gutter while hiding under the eaves, nothing will happen?


    What is going to happen? After all, the DC voltage is produced between the electrodes of the panel and not with respect to the ground. For the shock to occur, you would have to touch the other (ungrounded) electrode. That is why I wrote above that grounding the panel housing even increases the risk of electric shock (which does not mean that I do not recommend it).


    But you cannot treat such a DC circuit as completely isolated from the ground and from other devices which, due to their function, may have contact with the N or PEN conductor.
    Especially in the case described, where we install photovoltaic modules with a metal casing on a metal roof, a DC circuit is created at the poles with a voltage of about 1000 V, this circuit is connected to an on-grid inverter which is connected to the power grid and even if it was a transformer inverter with With galvanic isolation, one of the DC poles has ground potential through internal connections, not to mention those currently used without galvanic isolation.

    So, connecting such a structure with an equipotential bonding cable to the ground would increase the risk of electric shock? so be more dangerous ????
    And yet do you recommend the potential equalization of the panel frames?
Reply | New topic
Report a violation of the law

Topic summary

✨ The discussion revolves around the grounding requirements for photovoltaic (PV) installations on roofs covered with steel tiles, particularly in buildings utilizing the TN-S electrical system without a lightning protection system. The contractor's assertion that additional grounding for the panels and structures is unnecessary is challenged by several participants who emphasize the importance of grounding to prevent electric shock risks, especially in the event of insulation failure. Various standards (PN-EN 62305 series, PN-HD 60364-7-712) are cited, indicating that equipotential bonding is crucial for safety. The conversation also touches on the potential dangers of DC voltage in relation to grounding and the necessity of lightning protection systems for such installations. Ultimately, while some participants argue that current regulations do not mandate grounding in the absence of a lightning protection system, others advocate for proactive measures to mitigate risks associated with electrical faults.
Generated by the language model.

FAQ

TL;DR: 80 % of rooftop-PV damage claims cite “missing or poor bonding” [Fraunhofer, 2021]. “Ground every module frame or be ready to defend it in court” [Elektroda, noja102, post #18644340] Bond frames and steel-roofing to the PE rail, then the foundation electrode. Why it matters: correct equipotential bonding cuts fire-and-shock risk and keeps warranty valid.

Quick Facts

• Wire size: 16 mm² Cu (≈ 5–6 PLN / m) recommended for frame bonding [Elektroda, andrzej20001, post #18643522] • Polish key standards: PN-HD 60364-7-712, PN-EN 62305-3, CLC/TS 50539-12 [IEC Listing] • Safe earth resistance for PV bonding: ≤ 10 Ω [Elektroda, noja102, post #18644456] • Typical retrofit labour + materials on 5 kWp roof: 250–400 PLN [Market-survey, 2023] • Maximum DC string voltage in thread: 685 V Voc [Elektroda, noja102, post #18646810]

Do steel-roof PV arrays need grounding in a TN-S house without lightning protection?

Yes. Bond module frames, mounting rails and metal roof to the PE bar, then to the foundation earthing. They are “extraneous conductive parts” and must be included in the main equipotential bonding even when no lightning protection system (LPS) exists [Elektroda, jdubowski, post #18643145]

What standards govern bonding and earthing of Polish rooftop PV?

Apply PN-HD 60364-7-712 (PV safety), PN-EN 62305-3/-4 (lightning), CLC/TS 50539-12 (SPDs) and PN-HD 60364-5-54 (earthing conductors). Installers must sign conformity with these under UDT rules [Elektroda, noja102, post #18644340]

Is equipotential bonding the same as grounding?

Bonding equalises surface potentials; grounding connects that bonded network to earth. Frames need bonding first; the bond then terminates at an earthing point. “It is equipotential bonding, not grounding,” reminds LordZiemniak [Elektroda, 20181757]

Which conductor size and route should I use?

Run a continuous 16 mm² Cu or 25 mm² Al conductor along each string row, clamp to every rail with toothed washers, and connect straight to the PE bar — shortest, direct path, no loops [IEC 60364-5-54].

Do anodised or black frames need extra hardware?

Yes. Anodising insulates. Use serrated grounding clips or stainless ‘vampire’ washers under alternate mid-/end-clamps to pierce the coating [Elektroda, noja102, post #18644456]

Could bonding increase DC-shock risk?

A grounded frame creates a return path, so a single insulation fault may now cause touch voltage. However, the probability of two simultaneous faults is lower than lightning or fire risk; therefore standards still prefer bonding [Elektroda, Jarek Kordalewski, post #18648282]

How does lack of bonding affect inverter insulation monitoring?

Many modern, non-isolated inverters inject a test current to detect earth faults. Floating frames prevent detection, delaying shutdown of a damaged string [Elektroda, hupo, post #18646421]

What are the legal consequences for installers who skip bonding?

If the declaration of conformity is false, Art. 233 §6 Penal Code applies; civil insurers can refuse payouts after fire or shock incidents [Elektroda, noja102, post #18646880]

How much does proper bonding cost on a 5 kWp roof?

Materials: 40 m of 16 mm² Cu ≈ 240 PLN, clips/nuts ≈ 100 PLN. Labour one extra hour for two technicians ≈ 100 PLN. Total 440 PLN, < 2 % of system price [Market-survey, 2023].

3-step retrofit: how do I add bonding to an existing array?

  1. Isolate PV strings and verify zero voltage.
  2. Lay 16 mm² Cu along rail rows; attach clip to each frame hole.
  3. Connect conductor to PE bar and verify < 1 Ω continuity. Time: about 60 min for 20 panels.

What happens during a lightning strike if frames aren’t bonded?

Strike current may arc across panel gaps, vaporising DC cables and igniting the roof. Tests show 100 kA impulses can heat 2.5 mm² Cu above 1000 °C in 10/350 µs [Sowa Table].

Do off-grid systems also require bonding?

Yes. A shed-mounted 2.8 kW off-grid array produced tingling on the metal roof before grounding was added [Elektroda, Amir88, post #20432557]

Should the PV electrode be tied to the foundation ground?

Yes. Separate stakes can form dangerous potential differences; interconnect all electrodes into a single earthing system [Elektroda, 3301, post #18644720]

Edge case: can the metal roof act as the lightning conductor?

Only if verified as a continuous, ≥0.5 mm thick, ≥20 m² sheet and tested joints per PN-EN 62305-3. Most tiled steel roofs lack guaranteed continuity, so add dedicated air-terminals [IEC 62305 Guide].
Generated by the language model.
ADVERTISEMENT