1 KW SMPS Project Discussion

Status
Not open for further replies.

MicrosiM

Administrator
Staff member
1KW diy SMPS Project Discussion and Development




Hello All Registered users at the diysmps web site.

Let me thank you all for joining this forums, really appreciate it.

This thread (Hopefully) will discuss a 1KW SMPS Project, Explaining the SMPS sections, and a brief description about each part.

In order to help us all understand how the switching power supply works. And that will be in lessons.

I am not a SMPS expert either guru, therefore help required from any one who can help me and others.

There will be a PCB, Transformer, Construction details. To help you make that SMPS, while we can't cover every detail in SMPS since its open world.

An oscilloscope is a necessary tool for SMPS development, since there will be lot of examples, BEFORE you can skip to the assembly stage!

So at least you should have access to one.

Comments on this thread are welcomed from all, positive or negative.

At least I am trying to make all of us assembly a SMPS, since these things are rare on the web

Attached is SMPS Block Diagram with SMPS schematic


 

Attachments

  • SMPS Overview.jpg
    SMPS Overview.jpg
    29.6 KB · Views: 888

Solcar

New member
I look forward to this project being successful. I like your MOSFET driver buffer stage and I also like the International Rectifier MOSFET driver line of IC chips.
 

Attachments

  • IMG_0682.jpg
    IMG_0682.jpg
    1 MB · Views: 164

MicrosiM

Administrator
Staff member
I look forward to this project being successful. I like your MOSFET driver buffer stage and I also like the International Rectifier MOSFET driver line of IC chips.

I have chosen this chip, since its easy to operate, inlcudes interanl OSC, also needs minimal components to operate. Can you help into transformer design? for that SMPS?
 

Attachments

  • Booster(5000).rar
    211.4 KB · Views: 185
  • BoosterRing(2200).rar
    193.5 KB · Views: 398
  • Flyback(6000).rar
    418.5 KB · Views: 188
  • FlybackRing(3000).rar
    197.2 KB · Views: 188
  • Inductor(5100).rar
    207.9 KB · Views: 183
  • InductorRing(4000).rar
    189.2 KB · Views: 175
  • ExcellentIT(7100).rar
    223.2 KB · Views: 230

Solcar

New member
Alas I have learned transformer construction by trial and error. I prefer toroidal cores and don't generally like to have to keep leakage inductance low. When I use bridge and half bridge approaches I use zero voltage switching or a close approximation of it so that more leakage inductance isn't as much of a problem.
 

MicrosiM

Administrator
Staff member
Alas I have learned transformer construction by trial and error. I prefer toroidal cores and don't generally like to have to keep leakage inductance low. When I use bridge and half bridge approaches I use zero voltage switching or a close approximation of it so that more leakage inductance isn't as much of a problem.

We need FORMULAS, to help all understand that thing, transformer design, at least to know the number of turns + wire type. We know no one will give it to us, BUT we will keep trying.

until we make it possible for all.

Also you said before you are 30 years into SMPS, you should have some formulas?
 

Solcar

New member
I just know that about 5 feet of maget wire on a ferrite core translates to a few hundred microhenries. The formulas are complicated, and I generally don't feel comfortable using them. So, I estimate how much inductance that I want on the primary, and then use the turns ratio formula to get the number of secondary turns. One thing to watch out for is having too many ampere-turns on the core as that will overheat it and maybe lower the efficiency of the power supply if the transformer enters saturation. In my opinion, the tranformer is the hardest part.
 

MicrosiM

Administrator
Staff member
A good protection circuit, (Current Sense) will prevent the transformer from saturation, if we set it at 10 AMPS let say. so it will shutdown the driver.

We need the basic formula that make us move forward , easy one, we dont need complex formulas, I looked at those, if I follow them I will never be able to make a SMPS!!!


They are very complex, we only need things that makes a prototype possible.

you have any tips for PCB design?
 

Solcar

New member
On the SMPS circuit board help, I don't get the reduced EMI from them because I don't use them on my projects, all of which are just for personal use. I use point to point wiring built around the heatsink-mounted MOSFETs.

Either way, the key is to make loops handling high frequency power have as little area enclosed by them as possible. Also, try to keep power origination and return paths close together and parallel. That way radiated fields tend to cancel out at a distance. Try to keep low level signals away from the power circuitry.
 
Last edited:

Solcar

New member
The basic transformer formula is v1/v2 = turns1/turns2. The secondaries generally need to handle current in inverse proportion to their voltage reduction factor.
 

MicrosiM

Administrator
Staff member
Solcar;165]On the SMPS circuit board help, I don't get the reduced EMI from them because I don't use them on my projects, all of which are just for personal use. I use point to point wiring built around the heatsink-mounted MOSFETs.


Can you show us some pictures?
 

Solcar

New member
The circuits I build are compact but relatively ugly. I haven't ever taken any pictures of them since they were just for personal projects. I can check into trying to get a picture of something. One thing is that the 74C14 (CD40106) IC is easy to wire point to point upside down, with components soldered from one gate pin to another. The MOSFET drivers attach to the MOSFET leads and the controller circuit, built around the 74C14 IC, attaches to the MOSFET driver. I think it is compact and simple.

I forgot to mention another basic formula for transformers. I use the observation that 5 feet of wire on a ferrite toroid gives a few hundred uH of inductance. Then, I use the formula L1/L2 = turns1^2/turns2^2. I use by thinking that if the number of turns on a transformer doubles, the inductance quadruples. If the number increases by 3 times, the inductance increase by 9 times.
 

boraomega

New member
We need FORMULAS, to help all understand that thing, transformer design, at least to know the number of turns + wire type. We know no one will give it to us, BUT we will keep trying.

In a few days I am suppose to start doing my version of Half-Bridge SMPS and I will try to provide all of you with all necessary formulas and explanations. Just please don’t hold me firmly for words “few days” because I am so very busy with many other things and activities. That is the part of electronic related to audio amps and I want to complete my WEB site with that area to make it more useful – so, I will do that for sure.

Bora
 

Solcar

New member
One thing that I like to do is to build some impedance protection into my transformers. That way, I can skip active current limiting. Concentrating on just voltage mode control makes feedback loop stability easier. Zero voltage switching is an important part of this approach since it allows recovery of the extra energy that gets stored in the transformer when doing this.
 

MicrosiM

Administrator
Staff member
One thing that I like to do is to build some impedance protection into my transformers. That way, I can skip active current limiting. Concentrating on just voltage mode control makes feedback loop stability easier. Zero voltage switching is an important part of this approach since it allows recovery of the extra energy that gets stored in the transformer when doing this.

Intresting Idea!
 

Solcar

New member
I might have gotten the idea of impedance protection from the way that technique is very often used in 60hz transformers. It is especially commonly used in wall warts.

In SMPS, it can be done by means of having enough leakage inductance in the transformer itself. Alternately, an additional series inductor can be placed in series with the transformer, probably the primary winding, or a smaller than normal coupling capacitor can be used between the MOSFET totem pole and the transformer primary winding. If the smaller capacitor is used so that it experiences a voltage drop to the power supply rail under overload conditions, a diode needs to be placed across the capacitor to prevent its voltage from exceeding the power supply on at least one of the rails to prevent the output circuit from resonating and building excessive peak voltages.
 

MicrosiM

Administrator
Staff member
Drawing

I might have gotten the idea of impedance protection from the way that technique is very often used in 60hz transformers. It is especially commonly used in wall warts.

In SMPS, it can be done by means of having enough leakage inductance in the transformer itself. Alternately, an additional series inductor can be placed in series with the transformer, probably the primary winding, or a smaller than normal coupling capacitor can be used between the MOSFET totem pole and the transformer primary winding. If the smaller capacitor is used so that it experiences a voltage drop to the power supply rail under overload conditions, a diode needs to be placed across the capacitor to prevent its voltage from exceeding the power supply on at least one of the rails to prevent the output circuit from resonating and building excessive peak voltages.


It would be good to show a drawing with this idea, is that some sort of current sense?
 

Solcar

New member
It's a passive approach. i can show it on my latest project. It's hard to follow this entire circuit, so you might like to just look at the charge transformer 4000uH primary and 1000uH secondary. It lost my post, and I forgot a lot of what else i had said.

This one uses just leakage inductance to do the job. The topology of the charger is pretty much zero voltage switching.
 

Attachments

  • ebike-controller-charger.jpg
    ebike-controller-charger.jpg
    66.9 KB · Views: 748

Sierra

Valve Lover!
I dont understand

It's a passive approach. i can show it on my latest project. It's hard to follow this entire circuit, so you might like to just look at the charge transformer 4000uH primary and 1000uH secondary. It lost my post, and I forgot a lot of what else i had said.

This one uses just leakage inductance to do the job. The topology of the charger is pretty much zero voltage switching.


I really dont understand any thing from that circuit.

What I look for here is some help into the SMPS half bridge.

Can you help in that?

Dont post useless pictures


Sierra
 

arif

New member
hey all...dont mind the stupid question, but why use IGBT's?..as far as im aware, even the newer IGBT's are only capable of doing 40kHz or so..correct?..though if you used a cascode stage, this could be overcome at the expense of EMI...also, have you considered using desaturation control for primary side current limiting?..would love to be involved in this project..ive been wanting to make something like this for ages
 
Status
Not open for further replies.
Top