Posted by hallerb@aol.com on December 15, 2010, 8:25 pm
> T. Keating wrote:
> > wrote:
> >> Martin Riddle wrote:
> >>>> Here is a link to the diagram:
> >>>>http://www.tech-man.com/gti/220vgti.jpg
> >>>> Jim Rojas
> >>> God, that's not going to work. You need the GTI, at least ONE, and
> >>> switch the input source of the GTI to whatever input you want.
> >>> Cheers
> >> I updated the circuit for possible 220VAC adaptation. I just need to
> >> build a 2nd unit to see if this works.
> >>http://www.tech-man.com/gti/220vgti.jpg
> >> Any comments?
> > You reused.. C1-C2, F1-F4, Q1-Q4, D1-D4
> > Won't fly as input voltage is too low. 3-50vdc into 240vrms, (424
> > vac peak).
> > No circuit to prevent both rails from connecting to same leg
> > simultaneously.
> > Huge amounts of harmonic noise and current spikes..
> > Never get past FCC nor the UL. Will radiate RF noise like.
> > Hot DC+ leg. Safety hazard.
> > Electrolytic Cap's will blow. Internal resistance will cause them to
> > heat up and explode. They don't like wide voltage cycles.
> Any suggestions on fixing it would be greatly appreciated.
> Jim Rojas- Hide quoted text -
> - Show quoted text -
I got SPAM from china on grid tie inverters today... wonder if they
are related somehow?
Posted by BobG on December 16, 2010, 1:13 am
I think grid tie inverters need <5% distortion, which someone said
needs at least 7 bits. If you want 1200 watts for example, thats about
10 amps. Lets say you have a 0.1 ohm current sense resistor between
the incoming line and the outgoing inverter. If you sample the line
and load side of this resistor at the peak with a differential amp
(you are measuring the instantaneous current across the sense
resistor), you can adjust the inverter amplitude to be a little
greater than the line voltage. If the diff amp out is more than a
volt, the current is more than 10 amps, disconnect. This is just a
theory dreamed up in my brain. It might be completely impractical for
many reasons that I dont know about.
Posted by Jim on December 17, 2010, 2:47 pm
wrote:
>I think grid tie inverters need <5% distortion, which someone said
>needs at least 7 bits. If you want 1200 watts for example, thats about
>10 amps. Lets say you have a 0.1 ohm current sense resistor between
>the incoming line and the outgoing inverter. If you sample the line
>and load side of this resistor at the peak with a differential amp
>(you are measuring the instantaneous current across the sense
>resistor), you can adjust the inverter amplitude to be a little
>greater than the line voltage. If the diff amp out is more than a
>volt, the current is more than 10 amps, disconnect. This is just a
>theory dreamed up in my brain. It might be completely impractical for
>many reasons that I dont know about.
My Sun 500G Chinese 500w inverters sample the line twice per second,
and make adjustments accordingly. I have a video on you tube showing
the internals, but alas, no schematic. However, the circuitry doesn't
look excessively complicated, but it works great.
Do a search on You Tube for WA5DXP, and there one other fellow has
opened
them up also in his video. Mine was a 12V unit fused for 40 amps on
input, and
the other fellows was a 24v unit fused for 20 amps on input.
I don't think it could be built for anywhere near what you have to pay
for a complete unit, the components would cost a lot more if you
could find them.
Lots of info on You Tube search "grid tie inverter". I'm pumping
800+ watts into grid all day, plan to add more panels when finances
permit. Some fellows are pumping 2 kw+ and running their meters
backwards.
Jim
Posted by Jim Rojas on December 17, 2010, 6:37 pm
Jim wrote:
> wrote:
>> I think grid tie inverters need<5% distortion, which someone said
>> needs at least 7 bits. If you want 1200 watts for example, thats about
>> 10 amps. Lets say you have a 0.1 ohm current sense resistor between
>> the incoming line and the outgoing inverter. If you sample the line
>> and load side of this resistor at the peak with a differential amp
>> (you are measuring the instantaneous current across the sense
>> resistor), you can adjust the inverter amplitude to be a little
>> greater than the line voltage. If the diff amp out is more than a
>> volt, the current is more than 10 amps, disconnect. This is just a
>> theory dreamed up in my brain. It might be completely impractical for
>> many reasons that I dont know about.
> My Sun 500G Chinese 500w inverters sample the line twice per second,
> and make adjustments accordingly. I have a video on you tube showing
> the internals, but alas, no schematic. However, the circuitry doesn't
> look excessively complicated, but it works great.
> Do a search on You Tube for WA5DXP, and there one other fellow has
> opened
> them up also in his video. Mine was a 12V unit fused for 40 amps on
> input, and
> the other fellows was a 24v unit fused for 20 amps on input.
> I don't think it could be built for anywhere near what you have to pay
> for a complete unit, the components would cost a lot more if you
> could find them.
> Lots of info on You Tube search "grid tie inverter". I'm pumping
> 800+ watts into grid all day, plan to add more panels when finances
> permit. Some fellows are pumping 2 kw+ and running their meters
> backwards.
> Jim
I do agree that the small 250w grid tie units are cheaper to buy. But
the unit we are working on costs as much as a 250w unit to build, but it
allows up to 3000w. A Chinese 3000w inverter will set you back $00+.
Jim Rojas
Posted by The Ghost in The Machine on December 22, 2010, 9:59 am
> Martin Riddle wrote:
> >> Here is a link to the diagram:
> >>http://www.tech-man.com/gti/220vgti.jpg
> >> Jim Rojas
> > God, that's not going to work. You need the GTI, at least ONE, and
> > switch the input source of the GTI to whatever input you want.
> > Cheers
> I updated the circuit for possible 220VAC adaptation. I just need to
> build a 2nd unit to see if this works.
> http://www.tech-man.com/gti/220vgti.jpg
> Any comments?
> Jim Rojas
YOU'RE INSANE...NIX THE GRID TIE OR DROP THIS PROJECT IMMEDIATELY.
SANE PEOPLE ARE LIGHTING CHRISTMAS TREES AND THEIR HOUSES THIS TIME OF
YEAR.
I WOULD HATE TO HAVE TO RECOVER YOUR INSIPID SOUL FROM YOUR CHARRED
REMAINS.
PATECUM
XML site map
> > wrote:
> >> Martin Riddle wrote:
> >>>> Here is a link to the diagram:
> >>>>http://www.tech-man.com/gti/220vgti.jpg
> >>>> Jim Rojas
> >>> God, that's not going to work. You need the GTI, at least ONE, and
> >>> switch the input source of the GTI to whatever input you want.
> >>> Cheers
> >> I updated the circuit for possible 220VAC adaptation. I just need to
> >> build a 2nd unit to see if this works.
> >>http://www.tech-man.com/gti/220vgti.jpg
> >> Any comments?
> > You reused.. C1-C2, F1-F4, Q1-Q4, D1-D4
> > Won't fly as input voltage is too low. 3-50vdc into 240vrms, (424
> > vac peak).
> > No circuit to prevent both rails from connecting to same leg
> > simultaneously.
> > Huge amounts of harmonic noise and current spikes..
> > Never get past FCC nor the UL. Will radiate RF noise like.
> > Hot DC+ leg. Safety hazard.
> > Electrolytic Cap's will blow. Internal resistance will cause them to
> > heat up and explode. They don't like wide voltage cycles.
> Any suggestions on fixing it would be greatly appreciated.
> Jim Rojas- Hide quoted text -
> - Show quoted text -