low pass lc filter design

[FvM]

Seems like you have a lower pearmeability core like FT140-61. But the flux may be still too high and cause considerable inductance drop with 220V/16 kHz across the inductor. Low frequent current at 350VA load would almost saturate the core. Unfortunately I don't have material data and can't calculate the exact behaviour, it's just a rough estimation. Personally I'm not using ungapped cores for inductors, gapped ferrite or low permeability iron powder would be a better option.

 

[--BawA--]

Fvm, but I didn't connect any load after filter.perhaps it was the reactance current which was responsible for saturation..can u plz tell me how to calculate reactance current of the filter.i have used 0.22uf box capacitor.
Also plz tell me which iron core will be suitable for my purpose.??

 

[FvM]

For simplifaction, I assume that the full PWM frequent voltage is applied to the inductor. Reactant current is then simply 16 kHz voltage divided by reactance ZL (=2*pi*f*L). To calculate the 16 kHz voltage, you need to consider the SPWM modulation method (2-level, 3-level).

I don't know which iron powder cores are available at your site. Saturation limits of inductors can be calculated based on material data from manufacturer datasheets, but in addition, you should measure inductor currents in the real design, using variable DC bus voltage from a HV lab supply or an adjustable transformer powering the input rectifier.

 

[--BawA--]

thanx FvM,,
but , can u please suggest me a core, in which i can make 5mh inductor and it can handle 4ampere current without saturating?
also i have one more confusion, is CLC filter consumes less reactive power than LC filter?

 

[FvM]

CLC would be good, if the PWM carrier is 16 kHz sine. Bu it's 16 kHz square wave and an input capacitor causes huge reactive currents at 16 kHz harmonics. Pure L, LC, LCL or LCLC are the reasonable options.

I'm not used to powder core deign. As said, I'm mostly using ferrites. The Epcos ferrite tool is helpful to calculate inductors.

Increasing the rated current to 4A will require a considerable larger core. I guess that a 2 A inductor can work with a EE/EC/ETD 35 - 40 mm core and 4A with about 50 mm size, both with large air gap.

 

[--BawA--]

Hi fvm
If I use etd35 core then should I wind the wire on bobbin or simply wind the wire on core without using bobbin.?
One more doubt, can we not use toroidal iron powder core coz u have said that " u can use gapped core or iron powder core..
Which option is best to make 5mh inductor according to you.

 

[FvM]

You can use iron powder core, but I won't tell you a design procedure because I don't use these parts.

 

[--BawA--]

Thanx alot fvm
I will use ferrite core .. but please clear my 2 confusions
1) Which ferrite core should I use I.e gapped toridal core or gapped etd core ..?

2) if I use etd 35 gapped core then should I use the bobbin to wind the wire or simply wind the wire on core only?

 

[FvM]

Gapped toroids are technically possible, but I don't see any advantages. Iron powder cores can be considered as a core with built-in distributed gap. You can select between core types with different µr value, implemented by varying the iron powder to filler resin ratio. It works like an air gap.

For cores like ETD, bobbins are available and as I think, convenient to use.

 

[--BawA--]

Thanx fvm for your support
Now I will design my filter.
Finally last time I writing all the things which I have learned from u , please correct me if I am wrong anywhere.
Inverter ratings 220v , 350watts
Filter rating 5mh inductor on etd34 gapped core
0.22uf capacitor.
Cutoff frequency is 8khz.

Also suggest me a way to reduce the reactive current flowing in filter.?

 

[FvM]

Output filter design is always a compromise between conflicting design objectives:small size, low losses, low output ripple, low 50 Hz voltage drop etc. There's no universal design method. I think the suggested dimenioning is a starting point. You should calculate the results for your design and change the parameters if necessary.

 

[--BawA--]

ya thats right FvM ,but the problem is as i connect the filter all mosfets of my h bridge get damaged,,i dont get a chance to change the parameters, i have made it 6 times again n again, but the only thing i got was failure,, now i will again design my hbridge and new filter,,

 

[ahsan_i_h]

Perhaps, the part of the reactive current which returns to DC bus cant be completely absorbed by the DC bus capasitor (due to ESR and ESL )
As a result overshoot occurs at DC bus and IRF740 ( 400v max) fails.

So change mosfets to higher voltage ( 800V ) and connect more DC bus capasitors parallel and keep them close to mosfets. Then test again.
If it works then DC bus voltage needs to be observed by oscilloscope.

Also inductor's core must not saturate.

 

[--BawA--]

Plzz tell me any 800v mosfet that can handle atleast 2ampere and is easily available?

 

[ahsan_i_h]

TO-220 package:

IRFBE30 (800V, 4.1A)
IRFBE20 (800V, 1.8A)
BUZ81 (800V, 4.0A)
BUZ80A (800V, 3.0A)
2SK1030 (800V, 3.0A)

TO-3P package (bigger than IRF740):

2SK2611 (900V, 9.0A)

 

[--BawA--]

Hi ahsan.
Can I connect two 100uf electrolytic capacitor to my dc bus? Or some other capacitor

 

[ahsan_i_h]

Sure. Also add some ceremic capasitors ( 103 / 1kv are easily available ) two or three. Besides at no load your current consumption is
low, so you can use a 100w filament lamp in series with your main AC input for safety against short circuit.

- - - Updated - - -

New electrolytic capasitors are good.

 

[--BawA--]

Thanx ahsan.
I hve made circuit as u said . At low voltage test of my hbridge I.e 12.5 volt dc .I m getting almost 12v ac at the output. No load was connected at the output. I hve also seen the output waveform on DSO. and it was like 52hz square wave whose top was curve and distorted. It was not a pure sine wave
For filter I hve used 5mh inductor and 0.22uf capacitor..simple lc filter .
My carier frequency is 16khz .
Inverter rating 220v 50hz 350watt.
So where is the probem??

 

[FvM]

I hve also seen the output waveform on DSO. and it was like 52hz square wave whose top was curve and distorted. It was not a pure sine wave

It's the first time that you mention existence of oscilloscope measurements.
How about showing it?

 

[ahsan_i_h]

Hello Bawa,
I have seen two pdf files (from worchester polytechnic institute, WPI), these are

Dc/AC Pure Sine Wave Inverter
https://www.wpi.edu/Pubs/E-project/Available/E-project-042507-092653/unrestricted/MQP_D_1_2.pdf

PWM Techniques A Pure Sine Wave Inverter
https://www.wpi.edu/Pubs/E-project/...-190851/unrestricted/PWM_Techniques_final.pdf

In these pdfs one of the arms of H bridge is driven by 50Hz and other arm is driven by PWM signal.
But, perhaps, In your design top mosfets of the arms are getting 50Hz signal and bottom mosfets of the arms getting PWM
signal (but 10ms alternately). Is that right ?

If it is right, then your filter requires a minimum load current to work as expected.

can you send some link/pdf about your design ?

The pdfs are about spwm inverter of 120V AC. In page 36 of the second pdf (PWM Techniques A Pure Sine Wave Inverter) it is
stated that...

In order to move from theory to implementation, we needed to identify specific components for our design. First,
we needed to determine what MOSFETs we wanted to use. We know that in an H-bridge, the highest voltage a single
MOSFET that is off will experience is double the voltage of the DC power rail, which in our case is 2*170=340V.
We want to have a healthy safety margin above this voltage as well, with doubling the value being a good idea.
A rated voltage of 500-600V is therefore a good target for our MOSFETs.

I dont understand how 2*170=340V ??

As IR2110 is 500V, so 800V mosfet cant do any good right now. We need 800V (or more) IR2110 similar IC for 220V experiment.
Yes, it is better not to give 220v AC now. Filter should work properly at low voltage (max 110V ac),then
later voltage can be increased.

Can you give some link/pdf ( like WPI ) about 220V spwm inverter made by IR2110 and 500v mosfets ?