Transformer vs transformerless inverters

[Ramjam]

I'm led to understand that the market is moving towards TL inverters?

As I understand it they are slightly more efficient and are less fussy about string confirguration, just wanting as big a string as possible?

TL inverters are silent and Transformers get noisy as time goes by and a transformer inverter may well be a noisy beast in 10 years time?

TL inverters are firmware upgradable and as such new firmware in the future may well make them yet more efficient?

TL inverters have better comms built in making it easier to connect portable displays and internet connections?

If you've been wrestling with the TL possibilities what are you thinking?

 

[mmccx]

Just to note, the array frame will need to be bonded to earth on a TL inverter on a new earth spike in the earth system is TNCS. Also where not specifically stated the inverter will need to be protected by a class b RCD which are expensive and hard to come by.

 

[Ramjam]

Just to note, the array frame will need to be bonded to earth on a TL inverter on a new earth spike in the earth system is TNCS. Also where not specifically stated the inverter will need to be protected by a class b RCD which are expensive and hard to come by.

What do you mean by this, I need a special DC RCD?
If the supply is TNCS which most are in this area then I need to earth the frame to a dedicated earth spike as well? Surely its a dedicated spike or an RCD not both?

If there is no electrical separation between the AC and DC side then does the AC RCD not protect it?

 

[SolarCity]

What do you mean by this, I need a special DC RCD?
If the supply is TNCS which most are in this area then I need to earth the frame to a dedicated earth spike as well? Surely its a dedicated spike or an RCD not both?

If there is no electrical separation between the AC and DC side then does the AC RCD not protect it?

The reason you need the earth rod is because you shouldn't be bonding equipment outside of the equipotential zone on a PME system - although you may be able to speak to the DNO and get permission.

The class b RCD, as far as I am aware, is to prevent nuisance tripping. I could well be wrong though.

 

[dansk]

Class "B" prevents an injection of DC voltage into the grid, AC RCD dont pick up DC.

 

[brian wright22]

I have a system installed by EAGA.It has 3.54kw of Zen PV panels and a Fronius IG 3.6TL Inverter.
I have noted that the PV mounting/fixing frames are not earthed.I have questioned this with HomeSun & Eaga and initially did not get any sensible answers from them as to why there was no safety earthing of the PV frames.I am not in the trade but have been involved with Electrical safety & the BSI standards
I have had confirmation from Fronius "verbally I might add" that using this Inverter the PV metal mounting frames need to be earthed.
There is every possibility in my installation that if cleaning of the PV panels was required any time in the 25 year life span, contact could be made between the frames and a proper earth ,either from a ladder from an adjacent flat roof and the earthed TV aerial pole.
I have carried out a DC voltage measurement between the +ve and -ve conductors from the PV array before they enter the Inverter and the Fronius case.At the time when the system was producing 2.45kw the +ve was 354 volts & the-ve was 164 volts relative to the house earth.
Should a fault occure whereby either the PV +ve or -ve shorted to the PV frame (but not both at once) then this would elevate the frame to one of these voltages measured but not shut down the system .This would be relative to real bonded earth.
If the PV mounting frame was earthed then the system would shut down under this fault condition.
As far as I can deduce, all of the safety standards are quite clear from their decision trees as to when you should earth the PV mounting frame.
The Fronius IG 3.6TL is a Class I device.

Today I am getting a visit from Eaga's TSO to discuss my fears on the long term safety of my system and I will let you know the outcome of this meeting.

Regards Brian

 

[SolarCity]

I'm not surprised, if I'm honest, that some contractors aren't bonding the array. It's out of sight and I suppose a lot of contractors see this as out of mind as well.

Regarding the figures that you have been getting, this does surprise me. I would have thought that you should be getting virtually zero voltage between the DC leads and the house earth. Or am I missing something?

Out of interest, I suppose they didn't bother with the Class B RCD either?

 

[brian wright22]

As far as I know there is no Class B RCD.

Regarding the voltage between the House earth and DC input from the PV array I am not sure what the current would be that could flow if you grabbed hold of either dc wire and earth. I'm not going to try it but I may get the wife to do it!!

I have had a meeting with the EAGA TSO this morning and some very interesting comments were made.
One is that the DTI is to issue an update for the earthing arrangements for the Transformerless Inverters regarding the mounting frames.
The other outcome from our discussions and my request is that Eaga look at the Fronius IG 3.6TL dc input voltages and apply an increasing current from an RCD tester between Fronius case (Earth) and the either +ve or -ve wires to see if the system isolates the dc input wires. This will then possibly remove the need to have the PV frame earthed.

I will report back regarding the Eaga tests.

 

[SolarCity]

Thanks Brian, should be interesting.

 

[BrummyGit]

I've just found this forum browsing for some information (look out for post asking for help shortly). My situation is very similar to Brian.

I also have a HomeSun & Eaga installed system using a Fronius IG TL 3.6 Inverter. My system was initially connected via the RCD in my consumer unit which caused it to trip in a few minutes night or day. This turned out to be an error. The Fronius manual clearly states that the inverter has it's own RCD internally which will protect the AC side. It also states that if local regulations require an RCD, a minimum of 100ma residual current must be allowed for each inverter. My system is running fine now it is connected to the non-RCD side of my consumer unit.

During my installation a separate earth spike was installed and wired to the inverter, but the installation team explained to me that the panel frames needn't be earthed as they couldn't be reached from the only Velux in my roof. They told me that they had checked this with their supervisor.

From what I read in the Fronius manual I am not alarmed about the RCD, I'm interested in the outcome regarding earthing of the frames.

 

[brian wright22]

I've just found this forum browsing for some information (look out for post asking for help shortly). My situation is very similar to Brian.

I also have a HomeSun & Eaga installed system using a Fronius IG TL 3.6 Inverter. My system was initially connected via the RCD in my consumer unit which caused it to trip in a few minutes night or day. This turned out to be an error. The Fronius manual clearly states that the inverter has it's own RCD internally which will protect the AC side. It also states that if local regulations require an RCD, a minimum of 100ma residual current must be allowed for each inverter. My system is running fine now it is connected to the non-RCD side of my consumer unit.

During my installation a separate earth spike was installed and wired to the inverter, but the installation team explained to me that the panel frames needn't be earthed as they couldn't be reached from the only Velux in my roof. They told me that they had checked this with their supervisor.

From what I read in the Fronius manual I am not alarmed about the RCD, I'm interested in the outcome regarding earthing of the frames.

The situation at the moment is that Eaga technical are going to do a test at my request, tho establish what happens to the Fronius Inverter when an increasing load is applied between either the +ve & -ve DC input from the PV array and the inverter case ( house earth).What needs to happen is the reference to earth and the DC PV input requires to be isolated and all of my measurements suggests there not.

 

[BrummyGit]

Thanks, I'll keep an eye on this thread for the outcome.

 

[Ramjam]

It occurs to me that more modern high frequency transformer inverters are the way to go. They have high efficiency, galvanic isolation between AC and DC sides and are rugged and reliable. Until a consistent, practical and believable installation practice for TL inverters emerges I think they are a bad plan.

You mention the earthing of the array frame, according to BS7671 you can't take the house earth outside the equipotential zone so an entirely separate earth circuit needs to be supplied which along with the DC RCD(as rare and rocking horse poo) means than any extra income from increased productivity has been consumed by the installer in increased materials and labour.

It seems as though Eaga are guessing their way through this!

 

[BrummyGit]

Certainly with my installation they have earthed the inverter using a separate earth spike, and the Fronius IG TL appears to have the DC RCD built in. The manual on the IG TL states:

The design and function of the inverter provide a maximum level of safety
during both installation and operation.
The inverter provides operator and equipment protection through:
a) RCMU
b) monitoring the grid

RCMU = Residual Current Monitoring Unit
The inverter is equipped with an RCMU according to DIN VDE 0126-1-1.
It monitors residual currents from the solar module up to the inverter grid
connection and disconnects the inverter from the grid when an improper
residual current is detected.
Additional residual current protection may be needed depending on the
protection system of the installation or the requirements of the utility company.
In this case, use a residual current circuit breaker with a release
current of at least 100 mA.

 

[SolarCity]

Be aware that it is only in PME systems (TN-C-S) where taking the main earth out of the dwelling is not allowed. However, you may be able to get DNO permission to do so.