Energy Matters Forums

[cws]

Hi ,as an installer trying to compete with the Mc Solar companies, the use of transformerless inverters seems to be the go, does anyone have any thoughts on the merits of one type over another.I always liked the idea of a great big transformer,it seems like a more robust design,but............?
Rob

[Joey]

Hi ,as an installer trying to compete with the Mc Solar companies, the use of transformerless inverters seems to be the go, does anyone have any thoughts on the merits of one type over another.I always liked the idea of a great big transformer,it seems like a more robust design,but............?
Rob

I have had both , now using a transformerless , Much quieter , Much less heat so more efficient .

Massive amounts of copper goes into each transformer type inverter when it simply is not necessary now that MOSFET solid state switching has become so robust and At last more affordable.
There was a long period where copper was actually cheaper to use than these how power mosfets that are now becoming common place.

Efficiency figures up around 98% just isn't possible even with the best transformers , now at last there is a way to utilize more of what our panels make instead of turning it into wasted heat .

[cws]

Thanks Joey, i am familiar with the pro's. I have service contracts with 2 major inverter manufacturers and since the inception of tranformerless inverters i have been to a couple that have failed ,this rate, although very small concerns me, as, up till now ,for the ten or so years that i have been doing this ,the failure rate for tranformer type inverters seems much less , i can't prove it, but the quality of the mains grid waveform may effect transformerless inverters more than their predecessors.Any thoughts?

[Joey]

Thanks Joey, i am familiar with the pro's. I have service contracts with 2 major inverter manufacturers and since the inception of tranformerless inverters i have been to a couple that have failed ,this rate, although very small concerns me, as, up till now ,for the ten or so years that i have been doing this ,the failure rate for tranformer type inverters seems much less , i can't prove it, but the quality of the mains grid waveform may effect transformerless inverters more than their predecessors.Any thoughts?

I think there may be some minor teething problems , It seems SMA have really got a handle on this new technology and tested it to destruction long before it was released to the public.
It will be worth manufactures persisting because it is so much technology.

I'm really not that sure that grid waveform is going to present much of a problem for components that can operate thousands of cycles outside of 50hz , In the past couple of months I haven't seen my 4000TL disconnect from the grid due to excessive fluctuations , This is definitely much more of a problem with transformer type inverters which are much slower to compensate for fluctuations.
Solid state can react within a single AC cycle and and make the necessary changes instantly.

As for inverter failures we can only hope that any defects present themselves during the warranty period , which I think is generally the case with most failures like this anyway.

Something we haven't seen yet is a huge price reduction due to there being no transformer ( most of the weight of a standard inverter ) , Solid State components are considerably less expensive which has not been reflected in the new inverter pricing.

[bradley.jarvis]

I think you'll find that most transformerless inverters actually use IGBT's not MOSFET's which are better at switching high current. I think that the high end cost relative to the component cost has been offset by the R&D that the companies making them have had to invest. In the long run the transformerless will become much cheaper. I am designing a 3-phase inverter which should have no trouble switching upto 600V 50A using IGBT's, the cost of component's(from Farnell and Mouser) is no more than $200(not including PCB).

Thanks, Brad

[Joey]

I think you'll find that most transformerless inverters actually use IGBT's not MOSFET's which are better at switching high current. I think that the high end cost relative to the component cost has been offset by the R&D that the companies making them have had to invest. In the long run the transformerless will become much cheaper. I am designing a 3-phase inverter which should have no trouble switching upto 600V 50A using IGBT's, the cost of component's(from Farnell and Mouser) is no more than $200(not including PCB).

Thanks, Brad

Hi Brad ,

IGBT's Vs MOSFET's , this is another one of endless debates like Intel vs AMD CPU's .

This is a nice read to help people understand in basic terms , both have the pro's and con's and both can be used in inverter technology .
http://www.irf.com/technical-info/white ... wisely.pdf

Building your own inverter sounds like fun , wish I had the time .

[bradley.jarvis]

Hi Joey,
You are correct Mosfets and IGBT's are both quite suitable and consideration for both must be made. I have read that article before while doing research for my inverter. I have ended up choosing IGBT's for various reasons.

I should have probably said that IGBT's are also another technology that is used for inverters, not just MOSFET's.

I don't really have that much time, but designing has been fun (I work with AVR embedded processors and the new xmega processors is perfect for my application. I am currently working on isolation). I can't wait to get to building and testing, that will be fun. It is for a motor control application, it will be a >3kW transformerless inverter (3 phase but I will be running it straight off ~80-100VDC, no voltage boosting required). Hopefully it is reliable, a good transformer would be too big and bulky for my application.

Thanks, Brad

[Joey]

Hi Joey,
You are correct Mosfets and IGBT's are both quite suitable and consideration for both must be made. I have read that article before while doing research for my inverter. I have ended up choosing IGBT's for various reasons.

I should have probably said that IGBT's are also another technology that is used for inverters, not just MOSFET's.

I don't really have that much time, but designing has been fun (I work with AVR embedded processors and the new xmega processors is perfect for my application. I am currently working on isolation). I can't wait to get to building and testing, that will be fun. It is for a motor control application, it will be a >3kW transformerless inverter (3 phase but I will be running it straight off ~80-100VDC, no voltage boosting required). Hopefully it is reliable, a good transformer would be too big and bulky for my application.

Thanks, Brad

Hi Brad ,

Would be fantastic if you could keep us posted on the project , it's not something many could even dream about doing .
Something that would interest you in your work( sorry people if this is off topic ) but have recently just purchased the new APR5000 DZ
http://okay.com.au/okayshop/product_inf ... 3urrinhj94

Ends up closer to 55K by the time you get all the extra's you need , but what a machine!

[bradley.jarvis]

Hi Brad ,

Would be fantastic if you could keep us posted on the project , it's not something many could even dream about doing .
Something that would interest you in your work( sorry people if this is off topic ) but have recently just purchased the new APR5000 DZ
http://okay.com.au/okayshop/product_inf ... 3urrinhj94

Ends up closer to 55K by the time you get all the extra's you need , but what a machine!

Hi Joey,
If I get it working I will post the project and results to the forum. Still a lot of work to go though.

That rework machine looks like a nice toy to play with, I have steered away from using BGA packages because there hard to work with, that machine looks like it would make it easy. Imagine the tiny boards that you could make!

Thanks, Brad

[arie]

I like your BGA tool a lot better than my Aoyue 710 but let me comment on your project.

To make a grid tie inverter is not that complex a project but let me tell you
to get that to pass AS47777 so it is legal to use is is near impossible as a private project.

one can pickup a cheap gridtie inverter for under 200$ from china but try get it to pass AS4777
I looked at the list of inverters that have AS4777 a very short list

Arie