A couple things to still consider is
1) capacitor placement - capacitors should be placed equal distances to the diode AND the output cables so that they share current as good as possible.
2) ALL wire lengths should be as short as possible. The power supply would preferably be done on a PCB or a copper plate breadboard with large fat traces on all high current nodes.
3) All wires and traces should be as short as possible
4) All wires should be twisted pair minimizing loop paths.

 

HEY TINDEL!! You got me just in time. No Beers yet buddy. I own these http://www.ebay.com/itm/151790914548?_trksid=p2057872.m2749.l2649&ssPageName=STRK:MEBIDX:IT

This is what I did earlier today and have since pulled it all apart. It's tough getting all them caps together on the board but I did it. Just left one of them with longer leads. Somehow I knew about the distance thing. errr maybe it was just a happy coincidence.

 

Please... I really do not know why I put it together backwards.

Also, this inductor doesn't specify polarity as other inductors I've purchased did. Does it matter with this one? https://www.digikey.com/product-detail/en/SRP1770TA-2R2M/SRP1770TA-2R2MCT-ND/5429726

Thanks again, John

 

That one cap will be useless in the circuit Far too much inductance compared to the others. Try to group them in a 3/2 bunch like bowling pins.

Polarity on an inductor is new to me. polarity shouldn't matter.

 

Well Tindel I haven't fired it up yet, but I finally got those caps worked out. Just couldn't get a good bowling pin config. Or at least I couldn't wrap my head around it. So here she is!

 

perhaps fire was a bad choice of words. Smoke at the first resistor and the mosfet is pretty warm. Only 0.6v between my V+ and the mosfet Drain. I forget what it's called, maybe resonant frequency ?? I can hear it though, high pitched. Zero volts coming out at the load end.

 

probing V+ (battery +) and NEG at diode source or drain I get 0.75v. It does go up and down with pwm up and down.

 

Okay, I see my mistake. Ground from mosfet should not connect to Neg of caps. removing that and grounding caps from battery makes it work.

 

Maybe your pwm was stuck high?

 

No. I made a big mistake when soldering up the circuit. Look closely at the pic I put up. From fet drain to inductor to diode anode to cap neg. My work did not match my schematic. That's a problem. remove the connection from anode to cap neg and it works. I'm pretty sure I've explained it properly, but then again I am learning as I go. I still have a long way to go.

 

Tindel you're an animal! This is working great. I think I'd really benefit from a real oscilloscope, but this is working way way way better than anything I've come up with in three weeks. I did move my diode to stop my load before button press, but you definitely kicked it's ass.

Thank You!

 

John -

Thanks for the positive feedback (pun intended). We don't get much affirmation that any of the advice we give actually worked. So I'm glad it worked out for you and that you let me know! A 50W supply is no joke and difficult to design properly without some prior knowledge. If you're happy, then I'm happy.

I'm not following you on the button press or the new schematic. Was the new schematic how you messed it up the first time?

If you're serious about electronics then a oscilloscope is worth it's weight in gold. If not, then I wouldn't bother. But now that you've solved a few difficult problems, maybe you have the addiction too.

Tinman

 

No the new schematic is how it works. Well the first works too but wires got crazy hot and the mosfet and resistors would smoke. The diode being after the indutor was letting the the charge stored in the inductor go in both directions out of the negative side. Power was able to travel left toward the mosfet or right toward the diode. Placing the diode between the mosfet and diode lets power flow out of the inductor negative side and only travel to the right toward the caps.

I know now that there is no positive and negative side of the inductor but I'm referring to pos and neg by how it's connected. So each time the mosfet would turn on it was a direct short from Pos to Neg and then in the mosfet off time the inductor discharge would come all the way back to the mosfet. Moving the diode is protecting the mosfet in both the on time and the off time.

I sure hope I've explained this at least close to well enough.

John

 

This is how it should be hooked up:

http://i.imgur.com/sdZno3q.png

 

I agree. I basically followed the schematic from the max1523 that I had built previously. all I can say is that following my first schematic makes the mosfet smoke and wires get super hot. moving the diode before the indcutor doesn't make the mosfet or resistors smoke and wires are so cool their stiff. perhaps the circuit would benefit from a diode before and after the inductor.

 

I agree. I basically followed the schematic from the max1523 that I had built previously. all I can say is that following my first schematic makes the mosfet smoke and wires get super hot. moving the diode before the indcutor doesn't make the mosfet or resistors smoke and wires are so cool their stiff. perhaps the circuit would benefit from a diode before and after the inductor.

But it doesn't make a boost regulator either. The positive voltage just passes thru the inductor to the load. The FET and diode don't do anything.
If the first one was wired right, it could be the pwm from the micro was to low a frequency or it was always positive. Either of these would make the current thru the FET very high.

 

I've changed it back to match the first schematic and I get nothing out, just melty wires. I put it back to shematic number 2 and it works great. Other than I have some questions here...

I have control over the output via arduino pwm. Trouble is I can't get it above 4.2v with no load. 3.15v with a 0.05ohm load. The frequency of the inductor says 100kHz, and the arduino slowest so far is 340Hz. The fastest is around 12mHz. Now the 12mHz is untested because I have no oscilloscope that can measure that high. What I do know is that I've spent days now getting the code right and it should be doing 12mHZ. No matter what I do as far as frequency, I can not achieve a higher voltage.

Am I missing something?? or did I just answer my own question?? looking at the response from RonV. Maybe the frequency is way to high??

https://www.digikey.com/product-detail/en/IRLR6225TRPBF/IRLR6225TRPBFCT-ND/2538170
https://www.digikey.com/product-detail/en/SRP1770TA-2R2M/SRP1770TA-2R2MCT-ND/5429726
https://www.digikey.com/product-detail/en/RR71C101MDN1/493-3716-ND/2207252
https://www.digikey.com/product-detail/en/VB20100S-E3/8W/VB20100S-E3/8WGICT-ND/1543677

 

I've changed it back to match the first schematic and I get nothing out, just melty wires. I put it back to shematic number 2 and it works great. Other than I have some questions here...

I have control over the output via arduino pwm. Trouble is I can't get it above 4.2v with no load. 3.15v with a 0.05ohm load. The frequency of the inductor says 100kHz, and the arduino slowest so far is 340Hz. The fastest is around 12mHz. Now the 12mHz is untested because I have no oscilloscope that can measure that high. What I do know is that I've spent days now getting the code right and it should be doing 12mHZ. No matter what I do as far as frequency, I can not achieve a higher voltage.

Am I missing something?? or did I just answer my own question?? looking at the response from RonV. Maybe the frequency is way to high??

https://www.digikey.com/product-detail/en/IRLR6225TRPBF/IRLR6225TRPBFCT-ND/2538170
https://www.digikey.com/product-detail/en/SRP1770TA-2R2M/SRP1770TA-2R2MCT-ND/5429726
https://www.digikey.com/product-detail/en/RR71C101MDN1/493-3716-ND/2207252
https://www.digikey.com/product-detail/en/VB20100S-E3/8W/VB20100S-E3/8WGICT-ND/1543677

Not sure how you get 12 Mhz pulses from an 8 Mhz. part, but we can save that for another time.
Set the PWM frequency to 30 to 40 Khz.
Make sure you have some load - maybe the 2.7 ohm one - or when it works the voltage will rise and turn the caps into popcorn.

 

Hey Ron, Well I guess I'm (even still) a bit confused about what clock, timer or frequency is actually pertinent to my 3.3volt pro trinket and what I'm trying to do with it. The datasheet says the 5v is 16mhz and the 3.3v is 12mhz. I'm thinking that's referring to the usb speed. What I do know is that when I use my pin3 to pwm it does so at about 750hZ, using pin10 it does so at 350hz. I don't seem to done enough reading to have enough of an understanding of the atmega328 datasheet to fully control the pwm frequency to get to 30 - 40 Khz. Been at it for a few days now, just not happening. My oscilloscope can only measure up to 20Khz though so I really don't know.

 

Hey Ron, Well I guess I'm (even still) a bit confused about what clock, timer or frequency is actually pertinent to my 3.3volt pro trinket and what I'm trying to do with it. The datasheet says the 5v is 16mhz and the 3.3v is 12mhz. I'm thinking that's referring to the usb speed. What I do know is that when I use my pin3 to pwm it does so at about 750hZ, using pin10 it does so at 350hz. I don't seem to done enough reading to have enough of an understanding of the atmega328 datasheet to fully control the pwm frequency to get to 30 - 40 Khz. Been at it for a few days now, just not happening. My oscilloscope can only measure up to 20Khz though so I really don't know.

I'm not a micro guy, but I bet if you start a new thread and ask someone here will know how.

https://sites.google.com/site/qeewiki/books/avr-guide/pwm-on-the-atmega328