DC/DC Converter Help

[ronsimpson]

except for R12, on pin 1, did not solder...
I have to get something figured out to solder that better... Do I have to put more paste on to put it in the oven again???

I usually do the resistors by hand and the ICs by oven. Doing the resistors; some times I use a iron and solder, some time I put down paste, then the resistor, then just touch each end of the resistor with a iron.

Should my design work, if everything is correctly connected???

I think the 12 & 6.8 will be good. The 33V needs a higher current inductor or a load of 1.5A or less.

 

[ronsimpson]

Many people disagree with me! Many people remember the old way of doing things where each part is hand soldered and "thermals" are needed. Things are not that way now. When you heat a board with a oven thermals are not needed.

Example one. Heat comes out of the IC to a large pad on the PCB. This heat needs to be removed. See the small trace in-between the heat pad and the "heat sink" area. This is just like a resistor. It keeps the heat from getting out and away. Also this might also be a connection to ground for the high current. You just added a resistor to the circuit.

See how I try to get the heat out.

You can see that components that go to ground are soldered to ground with a thermal resistor.
Green lines show how heat moves.
Blue line, make bigger for high current.
Can you see more red? for ground and heat sink.

You might also look at the layout in the data sheet.

 

[ronsimpson]

Under full load you will have about 7 to 8 amps peak making these two loops. (I do not want to argue about which way current flows) Current goies out of C13, through L3, into IC, out of IC and back to C13.
Also current is making the other loop. C13---L3---D3---C14 then back to C13.
Make all these blue traces large.

Do you really want 8A pk, 3A average going through these small traces?

 

[LukeKnepp]

Okay, you say with the right size inductor and traces, I should be able to get 3A at 33V, or about 100 Watts of power...
However, the current through the inductor will be much higher...
I think that for my application, I will have to go to a boost regulator controller with an external MOSFET, do you recommend a certain IC???

I will definitely keep in mind your design tips...

Thanks to everybody for the help!!!!

 

[ronsimpson]

do you recommend a certain IC???

No. Here is a source of good tools. TI.COM then go to power. They show many choices but here is one. The software is good for helping choose what parts are good. I simply filler out a form Vin=16, Vout=33, Iout=3 and here is it.
There are pages to get more information. Example they say the inductor current (average) is 6.7A and the P-P ripple is 2.3A and the peak current is 7.9A. This helps pick a inductor. Some time they give a part number.

 

[LukeKnepp]

Why is this one only putting out battery voltage???

Is the Mosfet correct?

Are the feedback resistors correct?

Please!!!

 

[ronsimpson]

Why is this one only putting out battery voltage???

Because it is not switching. Nothing is switching.

Are the feedback resistors correct?

no. The voltage on IC1 pin-2 is too high. You can find out why.

Where did you get this circuit?

 

[ronsimpson]

Not optimized, first try. A little unstable. Probably needs soft start.

 

[LukeKnepp]

YES!!!!!!!!!

 

[LukeKnepp]

Nothing is switching.

Why? Can I fix that on the first schematic in #26?

The voltage on IC1 pin-2 is too high.

What size resistors should it be??

Where did you get this circuit?

I have a board that I got the schematic for, and the only thing I did was change R12 & R14 from variable resistors to fixed at the correct resistances...

 

[LukeKnepp]

Thank you for the quick reply!!!!!!!!!!!

 

[ronsimpson]

1) The job is the IC is to make 2.5V on pin2 "FB". You need to make a voltage divider so 33V=2.5V You know how to do that. As drawn this circuit will not work. Very common for internet schametics.
2) Area in green X. This is not needed if the input voltage is less than 20V.
3) Your post #26 is missing R5. This is a "current mode PWM". It must see the current in Q1.
4) U3a is too many parts. I did not use it.

 

[LukeKnepp]

Okay, I am trying this circuit here...

Will let you know the results!!!!

Thank you!!!!!!!!!!!!

 

[ronsimpson]

Will let you know the results!!!!

No. You changed parts with out knowing how it will effect the circuit.
1) Q1. The B-E can not be pulled above 0.7 volts. IC1-Output will try to pull up to about 15 volts. The IC can pull about 1A and will kill Q1. Also Q1 B-E will short out the Gate of the MOSFET.
2) You changed the value of L1. This changes the response of the circuit. You need to have a inductor rated for 10 or 15 amps.

 

[LukeKnepp]

What do I have to change, or what???

I have it on no load whatsoever, and it still only puts out battery voltage...

Do I have to do something different for the RT/CT or how does that pin operate??

 

[LukeKnepp]

Q1. The B-E can not be pulled above 0.7 volts. IC1-Output will try to pull up to about 15 volts. The IC can pull about 1A and will kill Q1. Also Q1 B-E will short out the Gate of the MOSFET.

Do I have to use the N-Channel Mosfet instead of the NPN transistor, or does that not make a difference??

 

[LukeKnepp]

 

[LukeKnepp]

Does this work as the 0.047 ohm resistor????

 

[ronsimpson]

Do I have to use the N-Channel Mosfet instead of the NPN transistor, or does that not make a difference??

You can not drive a NPN Base with out a resistor. If you don't have a small MOSFET then remove the transistor. In my circuit (#29) increase C7 from 330p to 1000p to get about the same frequency. 100khz.

Does this work as the 0.047 ohm resistor????

I do not know. I was trying to get 0.5 volts peak across the current measuring resistor.

 

[LukeKnepp]

Does the same IPP072N10N3 MOSFET not work for the small MOSFET as well?

I tried just removing the transistor, and this is what I got...

And in my board, I am using the IPP072N10N3, because I had nothing else...

It seems to work perfectly in the simulator, so I guess it is something else that is incorrect??

I measured across the small ohm resistor and got about 30mV...
So I changed it to a 2.9 foot long 22 gauge wire that is supposedly 0.047 ohm, according to this website...

And it still only measures about 30mV across it???

Do I have to apply a load so that the wire heats up and starts acting as a resistor???

Thank you!!!!!!