600/800W 220~230v AC Inverter SMPS from 12v DC

Status
Not open for further replies.

Take a look at TDK e141 - "Ferrite for Switching Power Supplies" (www.tdk.co.jp/tefe02/e141.pdf). The document states that the ETD34, in a forward converter mode, at 100kHz, can handle maximum 321W. The older document (attached) specifies 271W. So it all depends on the specific core you are using. I personally used an ETD34 core (purchased locally) at upto 150W. Perhaps it is best not to go over, say 150W, without further information about the core, which you may not find. For upto 300W or a little more ETD39 could be used. It is locally available and quite cheap.

I have a Chinese-made 400W inverter, which uses 2 ETD39-based transformers. It uses push-pull topology. On the other hand, the common 170W ATX PSU that I have seen, use an EI33 core (half-bridge topology). So, it really depends on the core and since you don't have much information on it, it's better to stay on the lower side of power estimates.

Hope this helps.
Tahmid.
 

Attachments

  • e141.pdf
    746 KB · Views: 403
Last edited:

Hi sam walkura,
I think we can stack two cores to get a higher output. First as walkura mentioned it is better to start with a smaller voltage o/p.
regards ani
 


the core selection is based on frequency, current density, flux density, and type of supply( push pull, fly back, forward etc), if you are increasing frequency flux density has to be reduced for 100khz flux density will be around 1100 B-guss,
 

@Tahmid & picgak: At this stage, I'm okay with not exceeding 150w for this unknown core. But the recommendation of 150w is for which frequency? I want to start designing 12v to 325v bridge converter. Now, I need to know the primary & secondary turns calculation with the required gauge of the wires.


rajudp said:
the core selection is based on frequency, current density, flux density, and type of supply( push pull, fly back, forward etc), if you are increasing frequency flux density has to be reduced for 100khz flux density will be around 1100 B-guss,
Click to expand...
Could you please show the calculation of output power (FullBridge Topology) of ETD34 using the data provided in the datasheet (attached by Tahmid) for different operating frequency (50khz, 100kHz)? I'm planning to buy TDK ETD34 core.
 

ETD34 core area = .971cm2 and Aw=1.220cm2
as per my calculation it is around 265w at 50khz and
348 W at 100khz assuming current density is 4A/Sqmm
this calculation is for Half and Full-bridge
 

rajudp said:
ETD34 core area = .971cm2 and Aw=1.220cm2
as per my calculation it is around 265w at 50khz and
348 W at 100khz assuming current density is 4A/Sqmm
this calculation is for Half and Full-bridge
Click to expand...

Your calculated power (for 100kHz) is close to the value shown in the TDK ETD34 datasheet. Effective cross-sectional area is .971cm2 but what is Aw? could you please show the calculation with the formula?

 
Last edited:

there is no simple calculation for selecting core for this you can use the formula

P=0.707 x J x F x Ac x Aw x B x 10E-8

P in watts
J current density of coil normally 400A/Sqcm
F Frequency in Hz
Ac Area of core Sqcm
Aw Window area of core Sqcm
B Flux density is 2000 till 35khz

more than that you have to select the core material that operates at the selected frequency and have low loss
 
Reactions: sam781

    S

    sam781

    Points: 2
    Helpful Answer Positive Rating

Thanks a lot dear rajudp to share that I really expected

Here is my calculation -
J=400 A/cm2
B= 2000
Ac=0.971 cm2
Aw=1.22 cm2
So, for F=50kHz; P=0.707*400*50e3*0.971*1.22*2000*1e-8=335w
For F=100kHz; P=0.707*400*100e3*0.971*1.22*2000*1e-8=670w

Results are not equal to your calculated value in post#45. Whats wrong here?

Does current density and Flux density not dependent on core material? B=2000 till 35kHz so, what about at 50kHz and 100kHz?
 

current density depends upon your coil selection,
for 50khz i have taken B as 1600 and 100khz it is 1050 (1075 actual)
 
Last edited:

rajudp said:
current density depends upon your coil selection,
for 50khz i have taken B as 1600 and 100khz it is 1050 (1075 actual)
Click to expand...

Where can I get B value with respect to frequency?
 

Does flux density depend on core material? Can I use this chart for all type of core?
 

this is not the exact values but will work with most types but there will be a +/- 100 difference in chart because i used 10 points to draw chart
 

Marty Brown's "Power Supply Cookbook" can u give a link to get this book
 

What about current density? Can I use it 400A/Sqcm for all core material?
 

yes you can use 400A for all core , if you are increasing the current density you have to take care of the heat generated by the coil , you can go up to 1000/Sqcm if wire length is below 1meter
 
Reactions: sam781

    S

    sam781

    Points: 2
    Helpful Answer Positive Rating
From tha datasheet of TDK ETD34 (shared by Tahmid )
Core factor - C1
Effective magnetic path length -le
Effective cross-sectional area - Ae
Effective core volume - Ve
Cross-sectional center pole area - Acp
Minimum cross-sectional center pole area - Acpmin.
Cross-sectional winding area of core - Acw

I wanted to map this parameter with the parameter of your formula. Here, Ae is probably Ac but canot correlate Aw (is it Acw?)

 

i dont thing that is the value we want , since it is a standard core check with some other company like epcos
 

Attachments

  • acw.JPG
    84.7 KB · Views: 148
Last edited:

Status
Not open for further replies.

Similar threads

Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn more…