Old 175 Watt Sharp Modules dont seem to work with new Fronius Primo 5000

[francissabato]

Keep getting error code when attempting to wire 36 Sharp 175 watt modules to replacement inverter. Fronius says the Primo wont work with the sharp panels because they are internally bonded to negative. Anyone have a cure for this? will microinverters work better?

Thanks

Francis

[francissabato]

still working on the sharp 175 watt/ fronius primo 5000 problem spoke to an engineer who had me load 3 strings on MPPT 1 and only 1 string on MPPT 2 seemed to work for a couple of days then back to 475 error code any ideas anyone? the primo was installed to replace two sunny boy 2500s that quit after 14 years.

[francissabato]

maybe a different approach....I have an array of 4 strings of 9 each Sharp 175 watt modules. Cant make it work on a Fronius 5000 Primo. Will they work with another inverter or micro-inverters?

[silversaver]

I am not an expert, but your old inverter were transformer inverter vs. Fronius Primo 5000 transformerless inverter. The grounding method were different. I might be wrong

[francissabato]

silversaver...what do people do with arrays of older modules when their string inverter craps out?

[silversaver]

The simple answer is call SMA. You have their old SB2500, ask them what would be the best model to replace them with. I guess your so call "engineer" had no idea what he or she is doing.

Or call a local installer.

[sunnyguy]

Fronius says the Primo wont work with the sharp panels because they are internally bonded to negative.

Not sure how they would be "internally" bonded to ground. Usually a grounded array is grounded at the inverter, so removing the old inverter should remove the ground. Then it is a matter of determining whether it was a positive or negative ground and connecting appropriately to the new inverter. Of course you may need to check whether the rest of the system meets current code requirements as far as wiring and equipment grounds. But replacing isolated inverters with transformerless is fairly common. Supposedly there are some things in nec2017 that favor this conversion. It is also possible that degradation of the insulation characteristics of the old panels is causing them to trip the sensitive ground fault detectors on the new inverter.

[inetdog]

Not sure how they would be "internally" bonded to ground. Usually a grounded array is grounded at the inverter, so removing the old inverter should remove the ground. Then it is a matter of determining whether it was a positive or negative ground and connecting appropriately to the new inverter. Of course you may need to check whether the rest of the system meets current code requirements as far as wiring and equipment grounds. But replacing isolated inverters with transformerless is fairly common. Supposedly there are some things in nec2017 that favor this conversion. It is also possible that degradation of the insulation characteristics of the old panels is causing them to trip the sensitive ground fault detectors on the new inverter.

Some early design panels (and even some currently sold?) would suffer a performance degradation if the substrate of the panel was at a high offset potential to ground (as seen at the frame of the panel).
Some panels would only work well if the negative side of the panel was at or above ground potential. Others would work well only if the positive side of the panel was at or below ground potential.
These were referred to respectively, as negative ground panels and positive ground panels.
Except for the earliest days when you could safely assume that panels would never be put in series (i.e. always battery panels matched to the battery voltage and feeding into a PWM CC), it is not possible to actually connect one terminal of the panel to the frame ground. If the Sharp panels in question are actually built that way, then they are of very limited use these days. You might be able to find microinverter or optimizer that would work with such panels, but it would be a long shot.
And regardless of whether the inverter or CC is transformerless or not, you would NOT be able to put two or more panels into a series string. Most if not all optimizers still require that all but one of the panels in a series string be isolated from ground. (That is, neither of the two panel terminals are internally connected to the frame.)
Theoretically, there could be a micro that will work under those conditions, but I would not be optimistic about finding one.