[RE-wrenches] Inverter 1741 Listing Process

boB Gudgel boB at midnitesolar.com
Mon Oct 18 22:42:16 PDT 2010


  On 10/18/2010 9:53 PM, Matt Lafferty wrote:
> Thanks for the info Dan and boB.
> What production tests required to maintain listing? Do you happen to 
> know if there is a different regimen for micro vs string vs central 
> inverters?


You mean, a difference between the two when UL, ETL, CSA comes by for 
their pop inspection ??

Off hand, I don't  see why one inverter would be given any different 
kind of attention than the other
type.  The UL/ETL followup inspections are basically to make sure you 
are following the proper
manufacturing and testing processes such as, software version, 
verification that you are using
UL recognized  sources of parts and materials, hi-potting at the correct 
voltages, etc.


They are both grid-tie inverters so the testing processes are very similar.

  I'm sure that Dan will pick up on anything  important that I forgot.

boB


> I'm working on developing a detailed commissioning procedure. Detailed 
> yet generic. There will be at least two versions for inverters. One 
> for string and one for central. In my book, central inverters have 
> re-combiners (standalone or integral) and string inverters don't. I'm 
> trying to understand which functions are tested in the factory on 
> every unit so I can avoid unnecessary duplication by the commissioning 
> people.
> Thanks in advance for any light you can shed on the subject!
> */Matt Lafferty/*
> ------------------------------------------------------------------------
> *From:* re-wrenches-bounces at lists.re-wrenches.org 
> [mailto:re-wrenches-bounces at lists.re-wrenches.org] *On Behalf Of *Exeltech
> *Sent:* Monday, October 11, 2010 10:04 AM
> *To:* RE-wrenches
> *Subject:* Re: [RE-wrenches] Inverter 1741 Listing Process
>
> Matt,
>
> The 5-minute delay is verified in the group of inverters sent to the 
> test laboratory.  Then, as long as the software and hardware don't 
> change, it's presumed the delay in all production units meets the 
> value(s) measured in the tested units.
>
> An alteration in either hardware OR software can result in a unit 
> being required to completely re-test (as Bob pointed out).  NRTLs get 
> copies of the source code and can and do periodically compare their 
> copy as submitted with the original test units to the software being 
> programmed into the inverters during production to verify it's the same.
>
> NRTLs conduct unannounced "field audits" by simply showing up on site 
> and randomly selecting various aspects of the product for verification 
> -- including the software.
>
> The overall UL1741 certification process is extremely complex, very 
> time consuming, and quite expensive.
>
>
> Dan
>
>
>
> --- On *Fri, 10/8/10, boB Gudgel /<boB at midnitesolar.com>/* wrote:
>
>
>     From: boB Gudgel <boB at midnitesolar.com>
>     Subject: Re: [RE-wrenches] Inverter 1741 Listing Process
>     To: gilligan06 at gmail.com, "RE-wrenches"
>     <re-wrenches at lists.re-wrenches.org>
>     Date: Friday, October 8, 2010, 9:26 PM
>
>     On 10/8/2010 6:04 PM, Matt Lafferty wrote:
>>     Hola Wrenches,
>>     Does anybody know off the top of their head if the
>>     /"5-minute-wait-to-interconnect"/ function is tested on 100% of
>>     inverters produced? /(i.e. every single inverter is tested with
>>     AC & DC within the start parameters of the unit for at least 5
>>     minutes)/
>>     Thanks!
>>     Matt Lafferty
>
>     Good question.  I bet it's not 100% tested because if the software
>     does not
>     change, then they may just not wait the extra 5 minutes in order
>     to save money on testing.
>
>     Or, maybe they do a random sampling for this test.
>
>     Theoretically, it shouldn't matter as long as the software does
>     not change
>     and the hardware is tested enough in other ways, like, timers and
>     clocks etc.
>
>     BTW, Nowadays, there are two options for manufacturers of grid
>     interactive inverters....
>
>     And the UL spec has gotten more stringent.
>
>     One option  is that you have to have  the  code (software) blessed
>     by the NRTL in a separate process (another UL specification, (UL
>     1998 --Software in Programmable Components")
>
>     OR, if they don't go for that option, then if software changes
>     need to be done to the inverter, the inverter must (technically)
>     go through the UL1741 listing process all over again.
>
>     boB
>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org/attachments/20101018/aef1bce2/attachment-0003.html>


More information about the RE-wrenches mailing list