My booster mod

[L18]

Hello all,
I think it is allowed to show own projects here, and since this forum was very helpful doing this, here goes. 

It has 2x18650 batterys parallel, and voltage range is ~3.7-6.2V


I wanted to make a mod which can be adjusted without any tools, so I put that(quite large) potentiometer there.
That small black button is which fires the atty trough mosfet.

I don't know how I managed to have so much parts, that had to be installed there:


Well, they fit there, and there is maybe 1mm to spare so I can get the lid closed. 
Solderings are quite messy, because I was feeling a bit frustrated since this was a bit longer thing to put together than I thought, it took me about 4-5 hours to build it.  But all solderings work, and that's the main point.
Booster circuit is PTN04050C.

Here you can see the batteries, two 2400mAh parallel gives 4800mAh, that is enough for quite some time.



Voltage range:



This mod is working very well, although I noticed that with 2ohm attys, output starts oscillating after 5.2 volts, I don't know what is causing it? Breaktru do you have any idea how it effects if capasitors on input and output aren't low-ESR? Because i'm not sure if those are which I used, I just had them lying around. Are tantalums best for this?

[Breaktru]

Very nice mod. Thanks for posting it.

Describe Oscillating. Is this a whining or buzzing sound?
Cap values may be high ESR. Every capacitor has some ESR. ESR is the electrical resistances in series with the capacitor plates.
High ESR will effect output. Cheap caps will have high ESR.

Remember, this module has a max of 2.4 amps @ 5v. Thus 5.2v w/ 2 ohm atty = 2.6 amps.
Do you see a drop in voltage?
I would also check for cold solder connections if whining or buzzing is heard. This indicates the converter starving for voltage.

Anyone want to chime in here? Your opinion and or expertize are welcome. This question has come up a lot lately

[MidnighToker]

Not sure about the low ESRs affecting it or not.  I always threw any old cap on my boosters. (grabbed a bunch of electrolytics off ebay real cheap)

Just speaking from my past experience...haven't worked with the boosters in quite a while.  Boosters in general don't handle high amperage well and once you hit 5.2v with a 2 ohm you are pushing 2.6 amps.  If you put a 2.5 ohm on there you should be fine up to 6v.

[L18]

Aargh, I'm sorry, of course I typoed this...I meant to write 3 ohms, which actually measured with multimeter was 2.7ohms.

And I have a oscillosscope, which I used to monitor output under load. Voltage was good and smooth until 5.2volts, but if I tried to get higher volts output started to oscillate about from 3.5v to 4.6v. I just realized that I forgot to check what was battery voltage, I just measured it and it was 3.75volts. I think that I'll try and load the battery to full, and see if I can get better results.

I didn't use the same resistor values as you Breaktru, they are pretty much smaller. I would love to know more how that feedback pin works, but I can't find any info from TI.
If it has some kind of maximum sink/source current limit or something...

[Breaktru]

Did you use a Mosfet? what is that?
Blowing up the photo,,, I would check for a cold solder especially at the Pos battery tab.

What values did you use and where?

[L18]

Yes, there is a IRLI2203N mosfet, because that little button can't handle that kind of currents, so all current goes through that mosfet, which is controlled by that button.
I have a 4.7kohm potentiometer, which middle leg is connected to Vadj. On the other leg is 1.8kohm resistor which is connected to ground, and on the other leg is 1kohm resistor which is connected to Vout. I think the small values COULD be the problem, I should try to put it like in your schematic to see if it works better with those.

[Breaktru]

I see a possible problem.
Are you using a 1k ohm from Output to AdJ ? This may cause a problem being too low.
And the 1.8k ohm is in series w/ the pot?

Put a meter on the Drain of the mosfet and make sure you are getting full battery voltage going to the converter input. If you're not you may have to tweak the resistor from Drain to Source.

[L18]

Yes, there is full battery voltage on converter input, measured against drain of the mosfet.
So I think it is that 1kohm which is the problem...damn, I was thinking about that when I assembled this whole thing, but of course it didn't cross my mind that I should test it with load... 
I guess I'll just have to try it...I'll take it apart, some day. 
How did you solve those resistor values for your mods, just trying different values?

EDIT: It's connected like this... 


Hmm, actually I think I see the problem now when I posted that picture...am I correct if I say that the voltage in Vadj can't be bigger that Vin? I calculated voltages that voltage divider would cause to the Vadj, and it would rise above Vin eventually.

[Breaktru]

Remove the 1k off the pot.
Use the resistor values I used for a 3.7v to 6.1v range.
4.7k in series to the 100k pot
and 15k accross Pin 3 & 4.
Don't put the output to the pot.

[L18]

Remove the 1k off the pot.
Use the resistor values I used for a 3.7v to 6.1v range.
4.7k in series to the 100k pot
and 15k accross Pin 3 & 4.
Don't put the output to the pot.

Yes, I'll have to do that. It just that I am such of curious person, that I would love to know why that doesn't work with resistor values I choosed, does anyone know the explanation? Because without load voltage range is what it is supposed to be, but with load oscillation starts above 5.2V?

[CraigHB]

Breaktru is doing a hack on the feedback circuit to force the converter into a different output range than shown in the data sheet.  He's found values that work, I imagine through a lot of trial and error.

The circuit you used is likely causing a feedback problem for the controller which destabilizes the converter.  There's no way to tell why it's happening without a full circuit schematic which is not provided.  My guess would be too much sink current due to resistance values that are too low.

DC-DC converters are naturally unstable.  When working with them from the ground up, it takes detailed design work to stabilize them.  TI and other chip makers deal with that for you by offering complete switching regulator modules. 

In building my own converters, I've run into instabilities that occur only under specific loading conditions.  Other times, I've encountered instabilities that occur only under heavy load or only under light load.  It can be downright bizarre.  You produced one that occurs only under heavy load.

Once a converter is unconditionally stable, there's a very limited range in what you can get away with changing.  Breaktru has found something that works which is not an easy thing to do.  There's no option other than to use the circuit with the values he has posted.

[Breaktru]

Breaktru is doing a hack on the feedback circuit to force the converter into a different output range than shown in the data sheet.  He's found values that work, I imagine through a lot of trial and error.

The circuit you used is likely causing a feedback problem for the controller which destabilizes the converter.  There's no way to tell why it's happening without a full circuit schematic which is not provided.  My guess would be too much sink current due to resistance values that are too low.

DC-DC converters are naturally unstable.  When working with them from the ground up, it takes detailed design work to stabilize them.  TI and other chip makers deal with that for you by offering complete switching regulator modules. 

In building my own converters, I've run into instabilities that occur only under specific loading conditions.  Other times, I've encountered instabilities that occur only under heavy load or only under light load.  It can be downright bizarre.  You produced one that occurs only under heavy load.

Once a converter is unconditionally stable, there's a very limited range in what you can get away with changing.  Breaktru has found something that works which is not an easy thing to do.  There's no option other than to use the circuit with the values he has posted.

Well said Craig, well said. I couldn't have said it better. Yes a LOT of trial and error went into it.

[L18]

Breaktru is doing a hack on the feedback circuit to force the converter into a different output range than shown in the data sheet.  He's found values that work, I imagine through a lot of trial and error.

The circuit you used is likely causing a feedback problem for the controller which destabilizes the converter.  There's no way to tell why it's happening without a full circuit schematic which is not provided.  My guess would be too much sink current due to resistance values that are too low.

DC-DC converters are naturally unstable.  When working with them from the ground up, it takes detailed design work to stabilize them.  TI and other chip makers deal with that for you by offering complete switching regulator modules. 

In building my own converters, I've run into instabilities that occur only under specific loading conditions.  Other times, I've encountered instabilities that occur only under heavy load or only under light load.  It can be downright bizarre.  You produced one that occurs only under heavy load.

Once a converter is unconditionally stable, there's a very limited range in what you can get away with changing.  Breaktru has found something that works which is not an easy thing to do.  There's no option other than to use the circuit with the values he has posted.

Thank you for this information.
While basic electronics is familiar to me, booster circuits are new. It is interesting to understand them better. 
I put the same resistor values as Breaktru, and now it is working break. Thanks for your help.

[Breaktru]

  See L18, all my tweaking and experimenting was NOT for nothing. 

[L18]

See L18, all my tweaking and experimenting was NOT for nothing. 

It sure wasn't, thank you for all that work you've done with this.