Best voltage regulator for batter driven applications ... DC-DC or 'DCM' ?

Go To Last Post
65 posts / 0 new

Pages

Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hello,

 

I want to develop some battery driven applications that require a constant voltage but
may not have a constant supply. Whether the supply be solar, battery, or mechanical.
I was looking at developing a circuit like a 'joule thief' that would buck or boost and
be very efficient. Then I started looking at finding a chip to do all that work for me.
I found the following chip 'MC34063A', but as I read it does not seem to be the right
choice.

 

The package is probably what would work best: 8 pin DIP.
From reading on 'Joule Thief ' designs they can bump up the output voltage when the
source voltage drops to as little as 0.6v. The specs on chips I am finding don't see to
go that low.
I found some information on DC -to- DC converters and wonder if what I am looking for is
a 'DCM' chip ?

 

The source will most likely be a pair of 1.5v batteries in series for 3v. for starters

If someone can provide a link to a chip or even a tutorial on how to select a chip like
this would be helpful.

 

Any help is appreciated.
Thank You.

Thank you.
Sid.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Welcome to AVRFreaks.

 

Leaving aside the fact that the 34063 is ancient, the first thing you should do before selecting a chip is define your performance requirements. What input and output voltage(s), what load(s), what range applying to each, efficiency, do you have space limitations, quantity to be built, target price, operating frequency limitations, operating temperature environment, etc.

 

Ross McKenzie, Melbourne Australia

Last Edited: Sat. Oct 10, 2020 - 06:31 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

DCM - don't come monday. 

 

You tell us nothing about what current you expect from the power supply, nor what batteries. This is critical information.

 

You should do some reading about batteries - they usually don't have a linear discharge curve - if they get down to 0.6V, then most common battery technologies are depleted - you're not going to be able to extract much energy.

 

If the 'joule thief' does what you want, why look elsewhere?

 

As well, when talking to engineers - don't use the term 'best' - f you have zero constraints, then 'best' might be easy to determine. Unfortunately, there are constraints like size, cost, availability etc.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Don't know what you mean by "DCM".

 

LinearTech (now Analog Devices) has several chips that work with VERY low input voltages and from very soft input sources, such as solar cells. But, those typically don't work with voltages much higher than 5V. So you REALLY need to work from inputs as low as 0.6V?  What are your real requirements? Valuesoft listed a variety of specs, and those are the major ones. 

 

Joule Thief is a very interesting little circuit, but it is not good for much more than a few milliwatts of power. Everything is a tradeoff!

 

Jim

 

Until Black Lives Matter, we do not have "All Lives Matter"!

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Sid03 wrote:
... or mechanical.
Those are typically a relatively high impedance (design will be difficult)

Sid03 wrote:
... 'MC34063A' ...
has a follow-on.

Sid03 wrote:
... or even a tutorial on how to select a chip like this would be helpful.
A book (AoE)

Sid03 wrote:
Thank You.
You're welcome and may you create joy.

 


Vibration Energy Harvesting | e-peas

NCP3063: Boost / Buck / Inverting Converter, Switching Regulator, 1.5 A | ON Semiconductor

https://artofelectronics.net/wp-content/uploads/2016/02/AoE3_chapter9.pdf

[page 24]

CHAPTER 9

VOLTAGE REGULATION AND POWER CONVERSION

 

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

What conversion efficiency are you thinking 70%?  90% .... 30% ?

The chip you pointed to is bipolar, so right off the bat you have a large loss compared to a mofet that might only drop a 10-20 mv at low currents.   The circuit current is hugely important...how much can you afford to waste on the circuit itself?  When the load is light (say a few ma), the circuit might use more power than that, if you are not careful.

You need some specs on the needed power levels & power usage profile & operating timeline (hours?, days?, months?, years?).

 

The package is probably what would work best: 8 pin DIP.

It will avery unlikely be a dip. At most it will be SO-8 or SOTt-23-x, and many newer chips will be in some form of dfn packages. 

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Sid03 wrote:
a pair of 1.5v batteries (sic) in series for 3v.

The trouble with that is always that you are limited by the "weakest link" - one cell will be dead before the other, so you won't get full usage.

 

If you're considering DC-DC, look at ones that will work from a single cell.

 

Also removes the possibility of users putting batteries in the wrong way around so that they aren't in proper "series" ...

Top Tips:

  1. How to properly post source code - see: https://www.avrfreaks.net/comment... - also how to properly include images/pictures
  2. "Garbage" characters on a serial terminal are (almost?) invariably due to wrong baud rate - see: https://learn.sparkfun.com/tutorials/serial-communication
  3. Wrong baud rate is usually due to not running at the speed you thought; check by blinking a LED to see if you get the speed you expected
  4. Difference between a crystal, and a crystal oscillatorhttps://www.avrfreaks.net/comment...
  5. When your question is resolved, mark the solution: https://www.avrfreaks.net/comment...
  6. Beginner's "Getting Started" tips: https://www.avrfreaks.net/comment...
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Sid has not returned since his initial post a month ago. Perhaps this is now a dead/deserted thread.

Ross McKenzie, Melbourne Australia

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Sid has not returned since his initial post a month ago. Perhaps this is now a dead/deserted thread.

Apparently they've caught the joule thief.  cheeky 

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

and maybe the 'Hamburgler'!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0


https://www.bbc.co.uk/news/busin....

 

valusoft wrote:
Perhaps this is now a dead/deserted thread.

Ah - so it is.

 

frown

 

just woken up by one of  gchapman's little "announcements"

 

Maybe also time to wake up the old discussion, "should dead/deserted threads be automatically locked"

 

I note that Segger and TI's forums do have a flag to mark a thread as "abandoned".

Top Tips:

  1. How to properly post source code - see: https://www.avrfreaks.net/comment... - also how to properly include images/pictures
  2. "Garbage" characters on a serial terminal are (almost?) invariably due to wrong baud rate - see: https://learn.sparkfun.com/tutorials/serial-communication
  3. Wrong baud rate is usually due to not running at the speed you thought; check by blinking a LED to see if you get the speed you expected
  4. Difference between a crystal, and a crystal oscillatorhttps://www.avrfreaks.net/comment...
  5. When your question is resolved, mark the solution: https://www.avrfreaks.net/comment...
  6. Beginner's "Getting Started" tips: https://www.avrfreaks.net/comment...
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I don't see how this helps. I think a thread that has been going on for years and years is very interesting to read and compare how things have changed. How interesting would it be for someone in the future to read a century old thread and be able to participate.

 

I understand how the joule thief works but I want to become good at dimensioning the components to my needs before starting testing a setup. If anyone has a good source (web-page, book, thread) with more in-depth info please let me know.

Also, I have never used an electronics simulation program before. Could learning to use such a program help in dimensioning a joule thief? Should I invest the time to learn to use a simulator?

TO THE FINDER... THE ISLE OF KOHOLINT, IS BUT AN ILLUSION... HUMAN, MONSTER, SEA, SKY... A SCENE ON THE LID OF A SLEEPER'S EYE... AWAKE THE DREAMER, AND KOHOLINT WILL VANISH MUCH LIKE A BUBBLE ON A NEEDLE... CAST-AWAY, YOU SHOULD KNOW THE TRUTH!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

 help in dimensioning a joule thief

Designing a joule thief?  Use LTSPICE...it is free and you will be learning it soon. 

If you don't need simulation, only design, you can often use Excel. 

https://www.analog.com/en/design...

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

tellSlater wrote:
(web-page,
There are many brilliant designers who write articles; subscribe to a few or several of your favorite journals.

 

tellSlater wrote:
Should I invest the time to learn to use a simulator?
Yes though which one?

Very partial to LTspice though effort on margins isn't easy (by SPICE parameters)

 


Series vs shunt linear voltage regulation for small solar-photovoltaic power supplies - EDN by Stephen Woodward

 

Intusoft's Home on the Web: SPICE Simulation, Analog and Mixed-Signal Circuit Design Tools, Magnetics Transformer Design Software, and Test Program Software

due to

The Art of Electronics 3rd Edition | by Horowitz and Hill

Download a sample chapter

[page 20, middle of right colum]

APPENDIX J: SPICE Primer 1146

J.1 Setting up ICAP SPICE 1146 [zero price demo version, 20 parts per schematic, 6% of the models]

LTspice: Worst-Case Circuit Analysis with Minimal Simulations Runs | Analog Devices

 

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

LTspice is great, but it (and most spice versions) do not simulate (or don't simulate very well) things like BJT collector breakdown or base reverse breakdown. 

 

As a result, the simulators typically are not so useful for applications that use component characteristics that are not listed on the spec sheet. For example, many BJT spec sheets will tell you not to exceed some reverse base voltage but they do not tell you what happens if you do.

 

Jim

 

Until Black Lives Matter, we do not have "All Lives Matter"!

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 1

Thank you, I will try LTSplice. I started using falstad yesterday and got to a working prototype of what I want to make. I quickly realized its shortcomings in simulating using small time steps. I don't think I can tune a joule thief properly using it.

TO THE FINDER... THE ISLE OF KOHOLINT, IS BUT AN ILLUSION... HUMAN, MONSTER, SEA, SKY... A SCENE ON THE LID OF A SLEEPER'S EYE... AWAKE THE DREAMER, AND KOHOLINT WILL VANISH MUCH LIKE A BUBBLE ON A NEEDLE... CAST-AWAY, YOU SHOULD KNOW THE TRUTH!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thank you, I will try LTSplice.

Why do you need it exactly (what are you trying to find out)?  Have you done the basic calculations already? You want to have a basic design ready & simulation has its own bag of stuff to contend with, especially when it comes to micro-parasitics (the real world may differ greatly).  For example, an opamp might simulate its output very accuarately, but shown drawing 2ua when its really draws 25ua (or the opposite) & normally you might care less.  So now you have to examine all assumptions very very closely.

Should I invest the time to learn to use a simulator? ​​​​​​​--Certainly, but recognize it can be misleading in some of the finest details, especially with all the possible settings.

 

https://www.electronicdesign.com...

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I am a mechanical engineer and have little knowledge on how I could dimension a joule thief by hand. It seems such a complex AC device to try and solve on paper. If there is theory for doing so though, please share.

 

I need to as efficiently as possible extract energy from a 10f, 2.7V super cap, to power 6 white leds (DC 120mA max at 3-3.2Vf). It would be nice if it worked for a range of charge 0.5 - 2.7V of the capacitor.

TO THE FINDER... THE ISLE OF KOHOLINT, IS BUT AN ILLUSION... HUMAN, MONSTER, SEA, SKY... A SCENE ON THE LID OF A SLEEPER'S EYE... AWAKE THE DREAMER, AND KOHOLINT WILL VANISH MUCH LIKE A BUBBLE ON A NEEDLE... CAST-AWAY, YOU SHOULD KNOW THE TRUTH!

Last Edited: Tue. Mar 29, 2022 - 01:16 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

 It seems such a complex AC device to try and solve on pape

Where is your current schematic?

The simulator simulates your design & may help find values, but you need a general idea.  Otherwise the simulator is like driving all over without having any map. 

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0


Here is what I have come up with so far:

A two phase motor is used to charge the supercap. The two phases of the motor pass through two full bridge rectifiers in parallel. Their output voltage is clamped by a zener to not exced 2.7V supercap limit. The LED side is connected to the cap via a switch. The joule thief drives the leds through a filter. The leds(2x 5050 white leds) draw 18mA (each single element - 18 x 6 mA total) when the cap is at 2.7V and about 1mA when the cap is at 0.5V.

TO THE FINDER... THE ISLE OF KOHOLINT, IS BUT AN ILLUSION... HUMAN, MONSTER, SEA, SKY... A SCENE ON THE LID OF A SLEEPER'S EYE... AWAKE THE DREAMER, AND KOHOLINT WILL VANISH MUCH LIKE A BUBBLE ON A NEEDLE... CAST-AWAY, YOU SHOULD KNOW THE TRUTH!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0


tellSlater wrote:
Their output voltage is clamped by a zener to not exced 2.7V supercap limit.
Zener impedance is inversely proportional to voltage; shunt regulator in lieu of a zener diode (precise, low impedance, greatly less operating current, easy to make stable)

 

The Art of Electronics 3rd Edition | by Horowitz and Hill

Download a sample chapter

[page 110]

Figure 9.92. TL431 adjustable shunt regulator–reference. The resistive divider in the application circuit on the right sets the “zener” voltage to 10.0 V.

 

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

?

 

What is the real goal of this project?

 

Are you trying to maintain lighting when the power source goes out?

 

Why do you need to obtain power from two power sources?

Does one typically fail, while the other still works?

 

NO.

Don't put your LEDs in parallel.

That is problematic in the real world.

 

The forward voltage on the LEDs is not exactly the same for each LED, and therefore they will fight each other, and draw different amounts of current, and have different light intensities, and generate different amounts of heat, etc.

 

If you want all of the LEDs on at the same time, then you have two other easy options.

 

One can put a small "load balancing" resistor in series with each LED.

This works well, but wastes a little bit of energy in the resistors.

 

These days one would typically "stack" the LEDs in series, and use a purpose designed LED driver chip to boost the supply voltage, drive the string of LEDs, and control the current through them.

 

There are many such chips available these days.

 

JC

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

You do not need one resistor of 40 ohm.

 

I think I should remove also 1mF, or put a very small one.

Each of leds should have 3.9 ohm in series.

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I will look into these propositions, thank you. I stacked 6 LEDs in parallel mainly because I have seen 5050 LEDs being used with one resistor (e.g. in 5050 LED strips). So I thought maybe I could get away with using one resistor for two 5050 LEDs.

 

The generator part is basically a two phase step motor - one power source. The purpose of this circuit is a small hand-cranked light I am trying to design.

TO THE FINDER... THE ISLE OF KOHOLINT, IS BUT AN ILLUSION... HUMAN, MONSTER, SEA, SKY... A SCENE ON THE LID OF A SLEEPER'S EYE... AWAKE THE DREAMER, AND KOHOLINT WILL VANISH MUCH LIKE A BUBBLE ON A NEEDLE... CAST-AWAY, YOU SHOULD KNOW THE TRUTH!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

 Their output voltage is clamped by a zener to not exced 2.7V supercap limit.

Sounds very odd...usually you are trying to not waste even a microwatt...the zener turns your energy into heat. 

Are you sure you are building an energy harvester?

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

tellSlater wrote:
The purpose of this circuit is a small hand-cranked light I am trying to design.
There's an organic appeal to hand-cranked power but our bodies are 20% efficient, 25% for the exceptional among us, and 50% theoretical max.

Crystalline silicon photovoltaic (PV) panels are slightly better than us at efficiency though satellite PV is relatively efficient.

Wind-driven generators are efficient here.

Super/ultra capacitors have a place.

 

https://solarmade.com/component/search/?searchword=satellite&searchphrase=all&limit=0 (Solar Made)

The Art of Electronics 3rd Edition | by Horowitz and Hill

Download a sample chapter

[page 118, bottom right]

9.12.3 Energy storage in capacitors

...

[page 119]

In quantitative terms these stored energies are dwarfed by those stored in batteries; but for some applications capacitors are just what you want. Among their other virtues, they have long lives, infinite endurance (charge/discharge cycles), the ability to be fully charged and discharged in seconds (or fractions of a second), and very high peak current capability (i.e., very low internal resistance, ESR). A storage capacitor, teamed with a conventional battery, can provide the best of both worlds: extraordinary peak power along with substantial energy storage.
...

[Ragone plot]

...

[jump battery for cars and light trucks]

 

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Simulations shows there is only significant current through the zener when the capacitor is at full charge and the breakdown Voltage of the zener is reached. When the capacitor is bellow 2.7 the current through the zener seems to be in the 100uA region while the cap charges at 80mA. That means there are 270uW of unwanted losses at worst on the zener. I think it is acceptable for what I want.

 

Having the energy go through the zener and dissipate as heat when the cap is at full charge is what I want. If someone keeps cranking the thing for hours it will save the cap from being overcharged.

TO THE FINDER... THE ISLE OF KOHOLINT, IS BUT AN ILLUSION... HUMAN, MONSTER, SEA, SKY... A SCENE ON THE LID OF A SLEEPER'S EYE... AWAKE THE DREAMER, AND KOHOLINT WILL VANISH MUCH LIKE A BUBBLE ON A NEEDLE... CAST-AWAY, YOU SHOULD KNOW THE TRUTH!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

tellSlater wrote:
The generator part is basically a two phase step motor - one power source.

I did not know a stepper motor could be used as a generator, I learned something today, or did I?

 

 

FF = PI > S.E.T

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

It  probably can for something small like this. I am using this motor which is cheap and comes with a gearbox. I had to do a small trick to convert it to a bipolar and after that it will produce 10Vrms (on each phase) when turned by hand on no load. When converted to bipolar it has 200Ohm and 44mH per phase.

 

I haven't tried it IRL yet but simulation shows it charging my 10F cap at 2.7V after 7' of "cranking" (providing 10Vrms per phase).

TO THE FINDER... THE ISLE OF KOHOLINT, IS BUT AN ILLUSION... HUMAN, MONSTER, SEA, SKY... A SCENE ON THE LID OF A SLEEPER'S EYE... AWAKE THE DREAMER, AND KOHOLINT WILL VANISH MUCH LIKE A BUBBLE ON A NEEDLE... CAST-AWAY, YOU SHOULD KNOW THE TRUTH!

Last Edited: Tue. Mar 29, 2022 - 07:54 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 1

 it will produce 10Vrms

 

That's good news.

But what is the output voltage, real world, with a load on the coils?

 

At the end of the day, the open circuit voltage is of interest, but it is the current flow that charges the cap.

 

You will have to do some real world testing!

 

JC

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

It all may work, and you will do the right thing eventually.
 
This means: remove 30 ohm, zener and other diode and 
install a decent Cap (switching) charger.
 
Anything else is a waste - a luxury that you can not afford.
Not, if this is powered by human.
 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Have you tried this motor?  With the gearbox, some can't be driven very well from the output, but it depends upon the specifics of the gears (you need to test it).

For example, a worm gear (not this style) very rarely can be driven from the shaft.

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Are there switching cap charger ICs? I didn't know. I will search for them.

 

It seems like there is an IC for everything nowadays that will do the job 1000x better than implementing it on your own.

TO THE FINDER... THE ISLE OF KOHOLINT, IS BUT AN ILLUSION... HUMAN, MONSTER, SEA, SKY... A SCENE ON THE LID OF A SLEEPER'S EYE... AWAKE THE DREAMER, AND KOHOLINT WILL VANISH MUCH LIKE A BUBBLE ON A NEEDLE... CAST-AWAY, YOU SHOULD KNOW THE TRUTH!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I worried this was the case but it turns out there are no worm gears in its gearbox. The ratio is great for turning it by hand.

 

It produces 10Vrms when turned by hand with medium effort (as much effort as swinging a door knob) on no load. It does have big phase resistance (200 Ohms) that will limit the output current though.

TO THE FINDER... THE ISLE OF KOHOLINT, IS BUT AN ILLUSION... HUMAN, MONSTER, SEA, SKY... A SCENE ON THE LID OF A SLEEPER'S EYE... AWAKE THE DREAMER, AND KOHOLINT WILL VANISH MUCH LIKE A BUBBLE ON A NEEDLE... CAST-AWAY, YOU SHOULD KNOW THE TRUTH!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I found this supercapacitor charger IC which is impressive but seems like overkill for my application.

TO THE FINDER... THE ISLE OF KOHOLINT, IS BUT AN ILLUSION... HUMAN, MONSTER, SEA, SKY... A SCENE ON THE LID OF A SLEEPER'S EYE... AWAKE THE DREAMER, AND KOHOLINT WILL VANISH MUCH LIKE A BUBBLE ON A NEEDLE... CAST-AWAY, YOU SHOULD KNOW THE TRUTH!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

tellSlater wrote:
I haven't tried it IRL yet but simulation shows it charging my 10F cap at 2.7V after 7' of "cranking" (providing 10Vrms per phase).

You should check with a scope, to confirm your simulations are in the right ballpark for charging.

 

tellSlater wrote:

I worried this was the case but it turns out there are no worm gears in its gearbox. The ratio is great for turning it by hand.

 

It produces 10Vrms when turned by hand with medium effort (as much effort as swinging a door knob) on no load. It does have big phase resistance (200 Ohms) that will limit the output current though.

How many of these do you plan to make ?

If you take a power-paranoid approach, you might want to load-match the generator, very like the MPPT ( Maximum power point transfer) controllers used in Solar Cell chargers.

Can the user even notice if the windings are shorted or open ? 

Last Edited: Tue. Mar 29, 2022 - 11:39 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Matching the winding resistance would mean I am getting the max power out of each winding, while minimizing the power dissipated as heat on them, right?

 

Easy to do if I was powering a resistor, but I have no idea how to make this supercap charging part of my circuit present a 200Ohm resistance to the motor. Any clues?

 

The motor shaft is about two times harder to turn when its windings are shorted.

 

I'll make a few prototypes for now.

TO THE FINDER... THE ISLE OF KOHOLINT, IS BUT AN ILLUSION... HUMAN, MONSTER, SEA, SKY... A SCENE ON THE LID OF A SLEEPER'S EYE... AWAKE THE DREAMER, AND KOHOLINT WILL VANISH MUCH LIKE A BUBBLE ON A NEEDLE... CAST-AWAY, YOU SHOULD KNOW THE TRUTH!

Last Edited: Wed. Mar 30, 2022 - 12:35 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

tellSlater wrote:
Matching the winding resistance would mean I am getting the max power out of each winding, while minimizing the power dissipated as heat on them, right?
No; maximum power is transferred when the source resistance is equal to the load resistance (Jacobi's law)

tellSlater wrote:
I'll make a few prototypes for now.
Cool!

 

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

but I have no idea how to make this supercap charging part of my circuit present a 200Ohm resistance to the motor. Any clues?

Ultimately, draw a certain current at the measured supplied generator voltage  (R=V/I), you can control the current drawn (different on each side) by varying the voltage applied on the charging side (for example, if the voltage were a perfect match, no current would draw on either side).

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

tellSlater wrote:

Matching the winding resistance would mean I am getting the max power out of each winding, while minimizing the power dissipated as heat on them, right?

Easy to do if I was powering a resistor, but I have no idea how to make this supercap charging part of my circuit present a 200Ohm resistance to the motor. Any clues?

 

Maybe trawl parts like Energy Harvesting, since that is what you are actually doing here.

https://www.analog.com/en/parame...

 

Some might suggest using a MCU with fast PWM support, if you have lots of time to debug things :) 

 

When into the final tuning mode, you can also improve your bridge rectifier, with MOSFETS bridge connected.

Simplest here is to choose mosfets with slightly better vgs ratings, it seems 25V and 30V gate rated parts are reasonably common, so you just need a spike clamp above that (2 zeners in series). 

Your 10V RMS will be inside that, but someone may flick it harder :) 

 

 

tellSlater wrote:

The motor shaft is about two times harder to turn when its windings are shorted.

That sounds promising, if you can feel a difference,  some useful energy transfer is occurring.

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 1

Of course you may be familiar with the LM3550

https://www.ti.com/lit/ds/snvs56...

there are prob some others

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

Last Edited: Wed. Mar 30, 2022 - 04:32 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

tellSlater wrote:
200Ohm resistance

 

This is comparable with my sealed reed-relays for 5V.

A tiny wire (can be 0.05mm), which is not good news to you... this 200 ohm is too much.

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I can control the voltage on the charging side by using a buck boost converter between generator and charging side, correct?

TO THE FINDER... THE ISLE OF KOHOLINT, IS BUT AN ILLUSION... HUMAN, MONSTER, SEA, SKY... A SCENE ON THE LID OF A SLEEPER'S EYE... AWAKE THE DREAMER, AND KOHOLINT WILL VANISH MUCH LIKE A BUBBLE ON A NEEDLE... CAST-AWAY, YOU SHOULD KNOW THE TRUTH!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

tellSlater wrote:
a buck boost converter

 

A step-down will be good enough.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0


So I was working on understanding why I need to use a buck converter and I did these simulations:

 

In the first one the capacitor is directly connected to the generator and it takes 5' to charge. Each phase generates 18V peak to peak AC and the voltage drop on the windings starts at 17V for an empty cap and when the cap reaches 2.7V the voltage drop on the windings is 15V.

 

 

Then I tried implementing a buck converter and attach it to the same generator:

 

The capacitor took 3' to cahrge up to 2.7V. I had to adjust the pulse width by hand as the cap was charging in order to keep the voltage drop on the windings to 9V, half the generated voltage. This should be my goal, right?

 

If my load was resistive, setting up one output voltage on the buck converter would be enough. Now that my load is a capacitor, the output of the buck converter should be adjusted according to the capacitor charge.

I have an MP2307 based buck converter board. Could there be a smart way to use its feedback loop in order to achieve this result IRL?

TO THE FINDER... THE ISLE OF KOHOLINT, IS BUT AN ILLUSION... HUMAN, MONSTER, SEA, SKY... A SCENE ON THE LID OF A SLEEPER'S EYE... AWAKE THE DREAMER, AND KOHOLINT WILL VANISH MUCH LIKE A BUBBLE ON A NEEDLE... CAST-AWAY, YOU SHOULD KNOW THE TRUTH!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

tellSlater wrote:
the voltage drop

 

Care with numbers, as I said in #34:

Anything else is a waste - a luxury that you can not afford.

Not a single Joule should be lost. You must have step-down regulator to transform the energy.
The law is known: U * I should be preserved after transformation.
Do not waste your Joules!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

 

 Could there be a smart way to use its feedback loop in order to achieve this result IRL?

 keep the voltage drop on the windings to 9V, half the generated voltage. This should be my goal, right?

The buck regulator already has feedback to create & maintain the output to whatever voltage you want, from whatever input voltage you are supplying.  Why is that a question ? You know the output voltage you want----set it, using 2 resistors 

note this chip is obsolete, pick a cousin.

 

 

 the output of the buck converter should be adjusted according to the capacitor charge.

 In general, NO.  The buck will work hard to bring up the cap using its max push.  However if it see the cap as a big-time dead-short, there might be some response mode, where it is designed to fallback, prevent overload, etc.

Many of these chips have adjustable current limit, soft start ramp up, etc that you should look into.

 

Your bigger issue is you have very limited input current, which can collapse the input voltage & halt the chip altogether.  So if the input voltage is going down, you want to force any current limit pin option to reduce the limit (though really you want to control the input current limit).  You can also get there by reducing the output setpoint (as input voltage falls pull FB pin up).  You'd prob be better off running some chip off its own battery, so it can transform the hand crank from zero volts up to its max voltage.  Most chips are not design to use an separate chip supply, so you'd look for one that can.

 

Before you get much further, you need to test your source and find out its capacity.  At 5V out, how much current can it supply, ....at 7V, at, 9v, etc.

 

The whole thing is sort of a mess.  You could use a simple Vreg or shunt reg, to only prevent the cap from getting an overvoltage, but that is very wasteful with high input voltages.

On the other end, with low cranking, you have almost no power and no voltage (or it simply drops like a rock), so running a chip is out of the question, and/or you need to limit the input current to allow the chip to work, without collapsing the crank voltage.  And whatever you do, if the chip does not have its own power source it will ride in & out of not having enough voltage (so perhaps you need a buck /boost instead, so the chip can bootstrap its own supply).

 

If you size the crank source so it can only output slightly more than the cap limit(say cranks at 3.5V max), it will inherently have more current capacity trade off at those lower voltages for a given size generator.  This can be supplied directly to the cap (perhaps via schottky diode)  & then it can charge the cap up all the way from zero & is self limiting.  At higher voltages (only), a shunt arrangement prevents cap overvoltage.  Since the crank can't exceed the limit by very much, there won't be much shunting happening & the efficiency will remain high. 

In other words, match the generator max to the cap & use a simple shunt to burn off any excess. The eliminates many circuit issues at low voltage, all the way down to zero.

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

Last Edited: Sat. Apr 2, 2022 - 04:20 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 1


 

surprise

Letting the smoke out since 1978

 

 

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Heh yeah, I needed to use that so I can keep the PWM at a low frequency and enable fast simulation times. In reality this inductor is much smaller but also the PWM is much faster.

TO THE FINDER... THE ISLE OF KOHOLINT, IS BUT AN ILLUSION... HUMAN, MONSTER, SEA, SKY... A SCENE ON THE LID OF A SLEEPER'S EYE... AWAKE THE DREAMER, AND KOHOLINT WILL VANISH MUCH LIKE A BUBBLE ON A NEEDLE... CAST-AWAY, YOU SHOULD KNOW THE TRUTH!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

You like simulation, then you can try a generator simulation.

 

Each current generator does have internal resistance, and the formula for efficiency is simple: maximum power delivered by the generator is when load resistance does match the generator resistance.

Try this simulation with 300 ohm and 100 ohm to see the power delivered to the load.

 

That is why generator must not have 200 ohm, but 2 ohm.

 

Last Edited: Sat. Apr 2, 2022 - 08:31 PM

Pages