MY 1KW SMPS Completed

Status
Not open for further replies.

chas11

New member
I was trying to make the point that a diy project should have the same design goals as a commerical product.
For a resonant supply the light load or no load is critical,some controllers go into skip cycle (onsemi NCP1395 & NCP1396) to protect components. Some supplies react different to a varying load as this changes the output impedance and then you have the classd amp which is a special case.Myself I use the same techniques that bench tech's use after they repair supplies,varying input ,loads and then burnin and retest and then ice test,heat test another burnin then an amp with rated wattage 20Hz,1kHz,and half fsw to see if compensation holds and then another sweep with network analyzer and scope for emi under loads and then if that goes well check notes and results for documentation.I am currently in the design and simulation phase of a LLC halfbridge converter which will use the NCP controller along with a PFC that onsemi has.
 

MicrosiM

Administrator
Staff member
Thank You Chas

I was trying to make the point that a diy project should have the same design goals as a commerical product.
For a resonant supply the light load or no load is critical,some controllers go into skip cycle (onsemi NCP1395 & NCP1396) to protect components. Some supplies react different to a varying load as this changes the output impedance and then you have the classd amp which is a special case.Myself I use the same techniques that bench tech's use after they repair supplies,varying input ,loads and then burnin and retest and then ice test,heat test another burnin then an amp with rated wattage 20Hz,1kHz,and half fsw to see if compensation holds and then another sweep with network analyzer and scope for emi under loads and then if that goes well check notes and results for documentation.I am currently in the design and simulation phase of a LLC halfbridge converter which will use the NCP controller along with a PFC that onsemi has.

Yes Chas.

I am just in the beginning of testing the first working SMPS I made.

I am not into your level in SMPS design, that's for sure

I wish that you support us here, and make me and other people learn

If you don't mind, please we need some help into the first diy SMPS project 1.

Sami
 

chas11

New member
I would think a good starting point is low input voltage since you don't have a pfc yet this will let you know the supply frequency range and also how well the components hold up,mosfets or IGBT's a scope can be used to find the lowest frequency the supply will handle.A good reference is appnotes AND8255-D and AN2450.It is very difficult to design for a wide input voltag range in high power LLC converters that is why a pfc is required to provide a constant bulk supply,one that varies about 50VDC under load.
chas11
 
Last edited:

MicrosiM

Administrator
Staff member
I would think a good starting point is low input voltage since you don't have a pfc yet this will let you know the supply frequency range and also how well the components hold up,mosfets or IGBT's a scope can be used to find the lowest frequency the supply will handle.A good reference is appnotes AND8255-D and AN2450.It is very difficult to design for a wide input voltag range in high power LLC converters that is why a pfc is required to provide a constant bulk supply,one that varies about 50VDC under load.
chas11


Thanks for the info.

One thing about the lowest frequency supply will handle, usually how do you mesure that?

But how about other amplifiers manufacturers, who build there amplifiers WITHOUT PFC, and the SMPS they have inside there amplifiers delivers up to 3KW constant, and 5KW music.

My next step is to work around the PFC. but this will occure after SMPS completes. 4KW & 6KW

Thanks
 

Dx Hugo

New member
Hello Renato.

You can make this supply, the other simple protorype of it. its not hard. its easy.

Can you get the parts for this?

IR2153D
ETD59
IRFP460
Ultra Fast Diodes

If you can access those. I will publish complete 800W RMS SMPS soon, Hard Switching, you can modify it to the voltage you want.

So if i find more people intrested into bulding it, I will do it fast..

Thanks

Hello MICROSIM,
I'm also interested in building the SMPS 800W RMS ...

I am impressed with your design, very good ...

Congratulations,

Dx Hugo.
 

MicrosiM

Administrator
Staff member
Hello MICROSIM,
I'm also interested in building the SMPS 800W RMS ...

I am impressed with your design, very good ...

Congratulations,

Dx Hugo.


Thank you, Please follow us on the diy SMPS project 1 for updates, I need to find more people intrested to start.

Can you get the parts required for this SMPS?

Thanks
 

chas11

New member
Sure it can be done as evident by designs as you point out, the basic principle is take a square wave ,wash it thru a LC filter and rectify the results to supply cuurrent to a complex load.The appnotes conver the frequency range of proper operation of LLC converters and by using the formulas you can calculate the frequencies of interest then measure the period of primary current or voltage(safely,ie isolate scope from ground),you might be able use a pickup loop close to a component and do the same for no load and max load.
 

Sierra

Valve Lover!
Sure it can be done as evident by designs as you point out, the basic principle is take a square wave ,wash it thru a LC filter and rectify the results to supply cuurrent to a complex load.The appnotes conver the frequency range of proper operation of LLC converters and by using the formulas you can calculate the frequencies of interest then measure the period of primary current or voltage(safely,ie isolate scope from ground),you might be able use a pickup loop close to a component and do the same for no load and max load.

Such tests are for regulated SMPS I think? :rolleyes:

And thats not a regulated SMPS.
 

chas11

New member
My post are in reference to LLC converter (basic),no regulation where currents for components are based on the correct calculations.You might want to refer to appnote AND8311/D,even if the supply is unregulated at some time during startup certain components are going to be exposed to large currents and voltage stress as an indication was when this supply was operated at a higher power output the power switches were hot indicating the load change moved the supply into a different mode of operation as discussed in numerous appnotes for the LLC type converter.If the users of this forum are LTspice users I have a couple of simulations that I can post if interested as far as the test I mentioned they can be used to test any type of supply,classd amps,motor controllers,closing feedback loops and others.
 

chas11

New member
Taking the reply in red with a grain of salt I figured we might be talking about two different types of supplies and after futher investigation it dawned on me that designs for SSTC's were being used for amps if that is the case SelfOSC HB drivers will present some basic problems.I am sure you have already have some ideas.Onsemi has some fixed freq controller IC's that might work well for your application,they include the mosfet drivers.You might need to change the mosfets as the IRFP460's have a large cgs.The only time I use free running is to chop the output of a pfc and since the pfc is regulated by feedback no control for the hb switch is needed.
 

MicrosiM

Administrator
Staff member
Taking the reply in red with a grain of salt I figured we might be talking about two different types of supplies and after futher investigation it dawned on me that designs for SSTC's were being used for amps if that is the case SelfOSC HB drivers will present some basic problems.I am sure you have already have some ideas.Onsemi has some fixed freq controller IC's that might work well for your application,they include the mosfet drivers.You might need to change the mosfets as the IRFP460's have a large cgs.The only time I use free running is to chop the output of a pfc and since the pfc is regulated by feedback no control for the hb switch is needed.


Any plans to show your SMPS here?
 

Dx Hugo

New member
MICROSIM, where is the design of smps 800w you promised to Renato and I?

We would like to build this project SMPS 800w ok?

Big Hug,

Dx Hugo.
 

MicrosiM

Administrator
Staff member
I am ready

MICROSIM, where is the design of smps 800w you promised to Renato and I?

We would like to build this project SMPS 800w ok?

Big Hug,

Dx Hugo.


Hello Hugo.


I cant remember that you asked me for that, however

No intrested people was found into this SMPS, therefore no progress was made.

I will supply you *here* with schematic and BOM, I am glad that you are still intrested into this, this makes me happy!

I dont know if you have access to the parts.


See attached schematic for SMPS, and tel me if you can get those parts.

Also the PCB is ready for this SMPS.
 

Attachments

  • 2KW_SMPS_IR2153.pdf
    36.4 KB · Views: 2,019

Dx Hugo

New member
MicroSim Yes, I would like to send me the pci layout SMPS IR2153 2KW! I really enjoyed the Sircuit looks pretty cool! Since already thank you for your help.
Please post for us here in the forum or send it to *my email*

Regards,

Dx Hugo.
 

MicrosiM

Administrator
Staff member
If possible also some information to make mount ok?:confused::confused::confused::confused:

I can send you the PCB files, but I have to modify it to work with ETD49 Transformer.

Becuase its mainly designed with smaller core, to save size on PCB.

Either you make that model or, you wait and buy a ready made transformer from the ones I manufcture. Becuase transformer is the most diffcult part into SMPS.

Because I am planing to make a group buy for this SMPS parts (Transformer + PCB) for all intrested, to enjoy making one for amplifiers.

But as I told you, no intrested people found. But I will help you make it.
 

Dx Hugo

New member
Very good ...
Could be used toroidal core? It would be easier and more efficient for me?

I would buy the parts "cores, FETs, IGBTs" right here "Brazil" avoiding as much as possible the import.

Big Hug, :cool:

Dx Hugo.
 

MicrosiM

Administrator
Staff member
Parts.......

Very good ...
Could be used toroidal core? It would be easier and more efficient for me?

I would buy the parts "cores, FETs, IGBTs" right here "Brazil" avoiding as much as possible the import.

Big Hug, :cool:

Dx Hugo.


Using toroidal cores, will make it difficult to wind and to isolate primary/secondary.

You need ETD49 Transformer, otherwise it will be very difficult to make this SMPS.

All other parts you can check in schematic, if you can get them or not.

Please tel me
 
Status
Not open for further replies.
Top