Can anyone recommend a DC-DC Power step-up converter (2v or so to around 18v output)

clonetwo

New Member
I'm currently using this XL6009 http://www.ebay.com/itm/10PCS-DC-DC-Adjustable-Step-up-Power-Converter-Module-XL6009-Replace-LM2577-/400543132967

I Don't know too much about my options but the XL6009 seems to work fine its just i don't like the working voltage range of 3v to 35v or so...

I need to power a 12 volt 1.5A regulator(L7812CV) from the 3 AA 1.5v pack so I think low working voltage would be nice also an output of 2v to 18v or so would work great.

Any ideas or modules any one can recommend would be of interest to me or myself, wait..i'm still here, ok good.

it might be the one i have is fine but i'm worried about voltage drop on worn battery pack. i know alkaline battery's pretty much bomb after 1v, so a 3v min operating voltage just might be fine.

Thanks

Regards,
Clonetwo
 

tmfkam

Senior Member
If you actually need 12V at anything like 1.5A you might struggle with this. 12V at 1A = 12W. To get 12W from 1.5V batteries you will need an 8A battery. If you are using 3 AA batteries in series (4.5V) then you will need a 2.66A battery. That will drain AA batteries very quickly. Even high capacity NiMh batteries will struggle to last an hour with that discharge rate.

Clearly if your 12V supply draws significantly less current than 1A, you might be able to generate this from AA cells. EPE did a design at one point using two AA cells to replace PP3 9V batteries...

June 2007 was the issue.
"3V to 9V DC Converter What a great idea! Never buy another 9V PP3/ 1604 battery again." http://www.epemag3.com/proj/0607.html

This was good for 9V at 90mA, adjustable for 4.5V - 20V out. Claimed to run for seven hours with 9V loaded to the 90mA quoted.
 

clonetwo

New Member

srnet

Senior Member
To get 12W from 1.5V batteries you will need an 8A battery. If you are using 3 AA batteries in series (4.5V) then you will need a 2.66A battery. That will drain AA batteries very quickly. Even high capacity NiMh batteries will struggle to last an hour with that discharge rate
Most standard Alkalines or NiMh will probably not be capable of that current in the first place.
 

westaust55

Moderator
If the 2N4401 transistor from your other post is adequate then that is rated at 609mA so you do not need anything like 1.5 Amps.
What is the actual current draw of your 3 LED strips?

Silicon Chip mag (Aust) has done a couple of 3 to 9 Vdc step up circuits over the years same as the EPE projects.
Built one (modified) some years ago.
Changing a resistor or two will allow you to step up from around 3 V directly to 12 Vdc but the chips in used in those magazine projects are only rated for around 50 to 90 mA at 9 or 12 volts.

More information on actual current draw will likely allow others to guide you better and may do away with the 12 V regulator as well.
 
Last edited:

nekomatic

Member
If you're using a 12 volt regulator it would make much more sense to generate a 12 volt supply to replace the output of that regulator, rather than generating 18 volts then immediately throwing away 6 volts, wouldn't it?
 

Goeytex

Senior Member
Yep, the (inefficient) 7812 linear regulator is really not necessary and could be eliminated from the circuit.
 

BeanieBots

Moderator
Yep, lose the 7812 regulator.
Also note that with a 3v input, the step-up PSU shown is only good for 0.4A at 12v. (limited by the 4A switching device)
You really do need to know (and let us know) the real power requirements.
 

nekomatic

Member
Is there any reason not to use a 12 volt battery pack as your primary supply, and step this down to 5 V for your Picaxe circuit? What kind of current do you need at 5 V?

Is there a minimum time that will elapse in between each 10-second requirement for the 12 V power?
 

tmfkam

Senior Member
Most standard Alkalines or NiMh will probably not be capable of that current in the first place.
Indeed not, I was more illustrating my point that the idea of generating 12V at 1A is unlikely to work particularly well.

I did find some mAh ratings for alkaline cells 'on the wild world wide web' (so may or may not be accurate). I had it in mind that quality AA alkaline cells might just reach 2000mAh but the figures quoted are: Standard Energizer AA - 1.5V, 2850mAh. So it may actually manage to work for a while. The losses due to internal resistance are going to be a major stumbling block as are the battery holders which probably will not be expecting to carry that amount of current, nor expect the heat generated by the batteries either.

Fortunately it would appear that 12V at 1A is not the actual requirement so the circuit might not have such a large battery drain as I originally thought.
 

tmfkam

Senior Member
At the risk of creating more work for CloneTwo...

I might be tempted to consider using a PWM output from the PicAxe fed into a switching transistor and inductor, powering the LED lights from the back emf, or using the PWM + transistor to drive a small HF transformer as a step-up and powering the LED lights from the transformer.

Could be more efficient?
 

Goeytex

Senior Member
At the risk of creating more work for CloneTwo...

I might be tempted to consider using a PWM output from the PicAxe fed into a switching transistor and inductor, powering the LED lights from the back emf, or using the PWM + transistor to drive a small HF transformer as a step-up and powering the LED lights from the transformer.

Could be more efficient?
I would not recommend this at all for someone relatively new to electronics and micro-controllers. Even experienced folks might find that to be a can of worms.
 

nekomatic

Member
I did find some mAh ratings for alkaline cells 'on the wild world wide web' (so may or may not be accurate). I had it in mind that quality AA alkaline cells might just reach 2000mAh but the figures quoted are: Standard Energizer AA - 1.5V, 2850mAh. So it may actually manage to work for a while.
Are you confusing cell capacity (in mAh) with current (in mA)? A 1000 mAh cell might well be able to supply 2000 mA, but it will run down in half an hour. The OP says they only need a few 10 second bursts of the 12 V power.

I still think stepping a 12 V supply down to 5 V will be more efficient and practical than the other way round though, unless there's a detail we don't know that rules this out.
 

tmfkam

Senior Member
Are you confusing cell capacity (in mAh) with current (in mA)? A 1000 mAh cell might well be able to supply 2000 mA, but it will run down in half an hour. The OP says they only need a few 10 second bursts of the 12 V power.

I still think stepping a 12 V supply down to 5 V will be more efficient and practical than the other way round though, unless there's a detail we don't know that rules this out.
I'm not confused, no. Possibly the person from whom the quote was taken might have confused the two terms, not me. I did say it was a direct quote and I have no way of verifying the accuracy of the statement. Originally it was implied that clonetwo was looking to provide 12V at 1.5A, from 3 AA batteries. That would need 4A. With your 100mAh battery 15 minutes would be the run time assuming the batteries internal resistance didn't become too high. Naturally in 10 second bursts it could be possible that the battery life would become much more acceptable.

Elektor in 2003 published a 'Mini Project' that generated 9V from a 3V supply. By changing the voltage regulator device (T3 in this circuit) for a 12V zener, close to 12V could be stored in a secondary capacitor and this might be able to be used to power the few second bursts? I found this link to the original Elektor article. http://www.digitalequilibrium.com/docs/9v_capacitor_battery.pdf

Trying to generate 12V from 4.5V I agree is much more difficult than going from 12V to 4.5V.

But a few HF pulses of 12V that look like a constant LED illumination might actually work out more efficient, if, as pointed out by Goeytex, not for a beginner.
 

clonetwo

New Member
OK, here are the AMP measurements for the circut.

power regulator is adjusted to 14v no load, under load it is 12.7v

My battery voltage was 4.3v

(I have set my multimeter to the mA setting.)
8.35 CD mA drwa on the positive battery wire for complete circuit. PICAXE on, relay on, and LED on and all npn transistors on, 315 MHz RX off.
1.45 DC mA draw from the positive of "XL6009 DC-DC Step-up power converter module".
1.35 DC mA draw on the positive LED wire of LED (after transistor(Q2) and 12v regulator (U2))

1.5 CD mA draw from boost with out regulator or transistor. the voltage is 12.5 no load 11.5 with load.


As i played with this it seems the the 12v regulator is NOT needed as the DC-DC boost drops below 12v and when i raise the voltage on the regulator the voltage under load stays below 12 volts.

what are your thoughts?

P.S. I found and ordered some other boost converters. here http://www.prodctodc.com/ultra-mini-dc-256v-to-412v-step-up-converter-3v-to-5v-1a-phone-charging-module-mobile-power-supply-p-204.html#.VCTPOKCojtt

Input voltage: DC 2.5V-6V
Output voltage:DC 4-12V (adjustable, adjust the fixed resistor)

I think they might be way better.


Thanks guys and sorry for taking so long to reply!!
Regards,
Clonetwo
 

clonetwo

New Member
At the risk of creating more work for CloneTwo...

I might be tempted to consider using a PWM output from the PicAxe fed into a switching transistor and inductor, powering the LED lights from the back emf, or using the PWM + transistor to drive a small HF transformer as a step-up and powering the LED lights from the transformer.

Could be more efficient?
i did do this before(i think) the project from this link https://learn.adafruit.com/diy-boost-calc/the-calculator, but i found the DC-DC boost module more efficient.
maybe there is a better way?
 

clonetwo

New Member
Is there any reason not to use a 12 volt battery pack as your primary supply, and step this down to 5 V for your Picaxe circuit? What kind of current do you need at 5 V?

Is there a minimum time that will elapse in between each 10-second requirement for the 12 V power?
the light will be fast BLINK, BLINK,BLINK,BLINK wait 1/2 sec and repeat 4 to 5 times. and will be activated by user and will not be used often 5 times a day i would guess.

I dont want to use a 12v battery because i need it to run from normal easy to get batterys, and i dont want the 5v regulator eating up the battery fast.
 
Last edited:

clonetwo

New Member
Are you confusing cell capacity (in mAh) with current (in mA)? A 1000 mAh cell might well be able to supply 2000 mA, but it will run down in half an hour. The OP says they only need a few 10 second bursts of the 12 V power.

I still think stepping a 12 V supply down to 5 V will be more efficient and practical than the other way round though, unless there's a detail we don't know that rules this out.
yes the 12v LED is the main function of the device but it will only be used maybe once a day to 5 times a day and blink for 5 to 10 seconds or so.

I was hoping that people can replace the battery easily when they die.
Though i am open to 12 battery that is cheap as AA and are easy to get. i have a 12v mini cell that powers my RX side of the circuit. and they are kinda expencive.
one day i might add a solar charger but that's a hole other deal.
 

nekomatic

Member
Those current measurements don't make sense to me. You're saying that when the LEDs are illuminated they only draw 1.35 mA from the 12 V supply, and the whole circuit including the relay only uses 8.35 mA? That seems far too low.

On the other hand, if it really is working as you want and those are the currents then it doesn't seem like you have much of a problem. Have you tried running it off two fresh batteries, to simulate the supply voltage you'll get from three nearly-flat ones?
 

tmfkam

Senior Member
i did do this before(i think) the project from this link https://learn.adafruit.com/diy-boost-calc/the-calculator, but i found the DC-DC boost module more efficient.
maybe there is a better way?
You surprise me. I've used similar circuits and got near to 80-90% efficiency. For the short bursts of light you are requiring I'd have thought that circuit the ideal answer, just 'blink it' with some short bursts of PWM, with the duty of PWM reduced to the minimum possible for the required brightness levels.

That particular circuit would require that the LEDs were connected in series, with at least four red LEDs in a chain so that 4 x 1.8V = 7.2V which is greater than the 6V supply. Otherwise the LEDs might still draw current with no PWM drive. This might have been the problem if your LED modules have driver chips on them? Different colour LEDs have different forward volt drops. I have assumed red LEDs, Blue/White LEDs drop around 3V so chains of 3 in series would suffice.

Multiple series chains of 3-4 LEDs could be paralleled to give the quantity or length of LEDs you require.
 
Last edited:

clonetwo

New Member
Those current measurements don't make sense to me. You're saying that when the LEDs are illuminated they only draw 1.35 mA from the 12 V supply, and the whole circuit including the relay only uses 8.35 mA? That seems far too low.

On the other hand, if it really is working as you want and those are the currents then it doesn't seem like you have much of a problem. Have you tried running it off two fresh batteries, to simulate the supply voltage you'll get from three nearly-flat ones?
Ya, I think they are off by a decimal point or two. that's the reading my multimeter gave me on the DC mA switch, other option is CD10A. i might try a different one and see what I get. I would really love to get my hands on an oscilloscope. :D

I'm going to try the 2 AA cells, thanks for tip i was trying to figure out how i could test low batteries.

thanks!
 
Last edited:

clonetwo

New Member
You surprise me. I've used similar circuits and got near to 80-90% efficiency. For the short bursts of light you are requiring I'd have thought that circuit the ideal answer, just 'blink it' with some short bursts of PWM, with the duty of PWM reduced to the minimum possible for the required brightness levels.

That particular circuit would require that the LEDs were connected in series, with at least four red LEDs in a chain so that 4 x 1.8V = 7.2V which is greater than the 6V supply. Otherwise the LEDs might still draw current with no PWM drive. This might have been the problem if your LED modules have driver chips on them? Different colour LEDs have different forward volt drops. I have assumed red LEDs, Blue/White LEDs drop around 3V so chains of 3 in series would suffice.

Multiple series chains of 3-4 LEDs could be paralleled to give the quantity or length of LEDs you require.
I dont know if more efficient is correct terminology. with the circuit on that link and pwm, I charged caps to get required brightness, but with the prefabbed dc-dc boost i got even brighter results and less wait time before it can blink again.

yes I'm using red LEDs it is a common 50-50 type strip. there are a total of 9 LEDs and there are resisters built into the strip as well.

I'm not sure how to figure out the LEDs specs(other then 12v) im gonna see if i can figure it out from data sheets. then ill know how much amps/mA it needs.

would it be a bad idea to hook the LEDs to a car battery and measure the LEDs amps to see the max draw of the LEDs? .. I just might try it as i do have some extra LEDS and worse case it will break the LEDs. I dont think it would damage the LEDs though...

I know this is a simple electronics circuit, but my lack of experience makes it hard to communicate or understand realistic amps values or other things.

Ill also test out the LEDs in parallel, the strip 'i think' is in series but i do have 3 sections here that i can a just to parallel. with 3 leds on each section and see if it is brighter.

Thanks,
Clonetwo
 
Last edited:

tmfkam

Senior Member
A lot of the PWM circuit would possibly need some experimentation with varying the drive frequency and duty cycle to determine the ideal combination for the LED sets you have that you are wanting to use. If the DC-DC module gives better results and offers acceptable battery life then there is not much incentive to try fiddling around to make incremental improvements.

I wouldn't have expected you to have to wait for long for the PWM circuit to charge up. The selection of the components is important though and could influence how well it worked, the transistor, the diode and the inductor should all be as specified by the calculator, with the PWM frequency set correctly too.
 
Top