Introduction: Adjustable Voltage Step-up (0.7-5.5V to 2.7-5.5V)
More by the author:
About: I like to experiment
Introduction:
A voltage step-up is a circuit that increases the voltage. It can be AC/AC, AC/DC, DC/AC or DC/DC.
This voltage step-up is a DC/DC adjustable voltage regulator. Usually a voltage regulator is fed by a higher input than output voltage, for example 9V IN to 5V OUT. This circuit will take a low voltage (down to 0.7V) and step it up to adjustable 2.7-5.5V. Since it is a regulator, the output voltage will stay constant regardless input voltage (0.7-5.5V), as long as output voltage is higher than input. It cannot step-down, only step-up. Are there any ICs that can do both?
I build it for usage in this project:
https://www.instructables.com/id/Battery-Charger-Powered-by-Fire/
Applications:
This is a typical circuit for a battery-powered USB-charger, for example two AA batteries (3V) to power 5V USB. There are tons of DIYs how to create that. They are often hard-specified to 5V output power. This construction can be used in a range of other applications. Many electronic devices work within 3-5 V and often you want to power them by low voltage power sources.
Some input examples:
- Solar panels
- Generators
- Batteries
- Peltier elements (TEC/TEG modules)
Some output examples:
- ICs
- Displays
- Motors
- LEDs
- Battery/USB charger
Costs:
All components cost me about 7€ but it's totally possible to find cheaper components!
If you want to construct this one, then continue reading!
Next Step: Specifications
Step 1: Specifications
I based this project on the Max757 IC. There are other DC-DC step-up modules but this one seems to be the best one for this purpose. It has high efficiency and low input voltage. If you want constant 3.3V or 5V output, then use Max756. Max757 is for adjustable output.
If you want less work and non-adjustable output, then you can buy a complete kit (version 3) with LT1302 IC at Adafruit. It´s also much smaller than my version. I bought it myself because it had higher output power than Max757, but I was not satified, I explain that in last step. If you need more power, then Max1771 seems to be nice (2-16.5V in and adjustable ?-12V out). It has a maximum power of 24W!
You can also buy an "emergency aa battery portable dc charger" on ebay and use the electronics in it. I bought one myself for about 1€ just for fun because it was free shipping from China to Sweden. That uses some kind of joule-thief I think, not very good replacement for this project but quite powerful compared to its size.
Full specification for max757 can be found in this pdf:
http://pdfserv.maximintegrated.com/en/ds/1167.pdf
Physical size of final board (WxLxH): 32x47x27mm (without inductor, height=17mm)
Next step: Materials
If you want less work and non-adjustable output, then you can buy a complete kit (version 3) with LT1302 IC at Adafruit. It´s also much smaller than my version. I bought it myself because it had higher output power than Max757, but I was not satified, I explain that in last step. If you need more power, then Max1771 seems to be nice (2-16.5V in and adjustable ?-12V out). It has a maximum power of 24W!
You can also buy an "emergency aa battery portable dc charger" on ebay and use the electronics in it. I bought one myself for about 1€ just for fun because it was free shipping from China to Sweden. That uses some kind of joule-thief I think, not very good replacement for this project but quite powerful compared to its size.
Full specification for max757 can be found in this pdf:
http://pdfserv.maximintegrated.com/en/ds/1167.pdf
- Minimum start-up voltage (@10mA load): 1.1V
- Minimum start-up voltage (@300mA load): 1.7V
- Minimum operating voltage (@20mA load): 0.7V
- Input voltage range: -0.3 to +7V
- Output voltage range: 2.7 to 5.5V
- Maximum output load (@input voltage=2V): 200mA@5V, 300mA@3.3V
- Maximum output load (@input voltage=1V): 50mA@5V, 75mA@3.3V
- Efficiency: Max 87% (depends on input voltage and output load
- Quiscent current (@no load, 2V input, 3.3V output): 60µA
- Operating temperature: 0ºC to +70ºC (Using Max75_C__)
Physical size of final board (WxLxH): 32x47x27mm (without inductor, height=17mm)
Next step: Materials
Step 2: Materials
I followed the Max757 PDF specification when selecting components. I also choose components that were in stock and at as low prices as possible. Keep in mind that there might be better choices! See image for the circuit scheme. I made some small modifications from the standard layout. I found out that pin1(SHDN) needs to be connected to pin6(OUT) in order to work (it need >1.5V i think). If it is floating the circuit will be very unstable. If it is grounded the booster is turned off.
Full specification of Max757 can be found in pdf:
http://pdfserv.maximintegrated.com/en/ds/1167.pdf
Inductor:
Should handle 1.2A (worst-case) and low DC-resistance (<0.03Ω) for good efficiency. I found one with 4.6A/0.031Ω.
Capacitors:
Should have low ESR (Equivalent series resistance) for good performance. Could not find any info about that. I choose one film capacitor (C3=0.1µF) and two aluminium electrolytic (C1=150µF resp. C2=100µF). I also added one (C4=1µF) film capacitor in parallel with C2 to filter high frequencies.
Rectifier Diode:
A switching Schottky diode is recommended for optimum performance. I choose 1N5817.
Resistors:
Does not matter I guess. The current will be very low. The output voltage is set by the equation:
VOUT = (VREF) [(R1 + P1) / R1], where VREF = 1.25V. Simplified to P1 = (R1) [(VOUT / VREF) - 1]
R1 and P1 should typically be 10kΩ to 200kΩ. I choose R1 to 47kΩ. I want 2.7V to 5.5V. Then based on R1, P1 needs to be between 55kΩ and 160kΩ. To achieve that I choose two resistors (P1R1 & P1R2), 47kΩ resp. 10kΩ (57kΩ would be equally good but I didn´t have that), in serie with a 100kΩ trim potentiometer. That means a range of 57-157kΩ and an output voltage range of 2.77-5.43V.
If you need low-battery detector (LBI-pin), then you would need two more resistors.
The threshold voltage is set by R3 and R4 using the following equation:
R3 = [(VIN / VREF) - 1] (R4)
where VIN is the desired threshold of the low-battery detector, R3 and R4 are the input divider resistors at LBI, and VREF is the internal 1.25V reference. R3 and R4 should typically be 10kΩ to 200kΩ. I skipped that part. See Max757 PDF.
I also put a red LED in serie with a 470Ω resistor (R2) to indicate OK/ON-state.
IC-Socket:
I recommend a socket for the IC. If it is destroyed it can easily be replaced.
Circuit board (Prototype board):
I used a pre-fabricated prototype board from an old project. That was not included in total price but it´s just 3€ for 100x160mm.
These can be found at farnell.com. Item number in brackets. I used other prototype board, resistors, potentiometer and terminal blocks but similar is listed here.
Tools needed:
Full specification of Max757 can be found in pdf:
http://pdfserv.maximintegrated.com/en/ds/1167.pdf
Inductor:
Should handle 1.2A (worst-case) and low DC-resistance (<0.03Ω) for good efficiency. I found one with 4.6A/0.031Ω.
Capacitors:
Should have low ESR (Equivalent series resistance) for good performance. Could not find any info about that. I choose one film capacitor (C3=0.1µF) and two aluminium electrolytic (C1=150µF resp. C2=100µF). I also added one (C4=1µF) film capacitor in parallel with C2 to filter high frequencies.
Rectifier Diode:
A switching Schottky diode is recommended for optimum performance. I choose 1N5817.
Resistors:
Does not matter I guess. The current will be very low. The output voltage is set by the equation:
VOUT = (VREF) [(R1 + P1) / R1], where VREF = 1.25V. Simplified to P1 = (R1) [(VOUT / VREF) - 1]
R1 and P1 should typically be 10kΩ to 200kΩ. I choose R1 to 47kΩ. I want 2.7V to 5.5V. Then based on R1, P1 needs to be between 55kΩ and 160kΩ. To achieve that I choose two resistors (P1R1 & P1R2), 47kΩ resp. 10kΩ (57kΩ would be equally good but I didn´t have that), in serie with a 100kΩ trim potentiometer. That means a range of 57-157kΩ and an output voltage range of 2.77-5.43V.
If you need low-battery detector (LBI-pin), then you would need two more resistors.
The threshold voltage is set by R3 and R4 using the following equation:
R3 = [(VIN / VREF) - 1] (R4)
where VIN is the desired threshold of the low-battery detector, R3 and R4 are the input divider resistors at LBI, and VREF is the internal 1.25V reference. R3 and R4 should typically be 10kΩ to 200kΩ. I skipped that part. See Max757 PDF.
I also put a red LED in serie with a 470Ω resistor (R2) to indicate OK/ON-state.
IC-Socket:
I recommend a socket for the IC. If it is destroyed it can easily be replaced.
Circuit board (Prototype board):
I used a pre-fabricated prototype board from an old project. That was not included in total price but it´s just 3€ for 100x160mm.
These can be found at farnell.com. Item number in brackets. I used other prototype board, resistors, potentiometer and terminal blocks but similar is listed here.
- IC (DC-DC step-up): MAX757CPA+ - IC, DC/DC UP CONVERTER, DIP8, 757 [1290852]
- Socket (DIP8): TE CONNECTIVITY / AMP - 1-390261-2 - SOCKET IC, DIL, 0.3", 8WAY [1101345]
- C1 (150µF): PANASONIC - EEUFR1H151, RADIAL, 50V, 150UF [1800679]
- C2 (100µF): PANASONIC - EEUFC1H101, RADIAL, 50V, 100UF [1848464]
- C3 (0.1µF): EPCOS - B32529C104J, FILM, 0.1UF, 63V, RADIAL [2112914]
- C4 (1µF): EPCOS - B32529C105J, FILM, 1UF, 63V, BOXED [1692372]
- D1 (1N5817): VISHAY FORMERLY I.R. - VS-1N5817, SCHOTTKY, 1A, 20V [1463212]
- D2 (LED): MULTICOMP - 703-0100, 5MM, RED, 400MCD, 643NM [2112111]
- L1 (22µH): PANASONIC - ELC16B220L, 22UH, 4.6A, 0R031 [1749170]
- P1R1 (10kΩ): MULTICOMP - MF25 10K, 10K, 0.25W, 1% [9341110]
- R1&P1R2 (47kΩ): MULTICOMP - MF25 47K, 47K, 0.25W, 1% [9341960]
- R2 (470Ω): MULTICOMP - MF25 470R, 470R, 0.25W, 1% [9341943]
- P1(100kΩ): TE CONNECTIVITY / CITEC - CB10LH104M, SIDE, 100K [1227527]
- Terminal block: IMO PRECISION CONTROLS - 20.101M/2, PCB, 2WAY, 22-12AWG [9632670]
- Prototype board: ROTH ELEKTRONIK - RE523-HP - PCB, FR2, STRIPES, 2.54MM [1172112]
Tools needed:
- Soldering iron
- Pliers
- Sharp object (knife)
- Small saw
Step 3: Pre-Test
I usually test my circuits before soldering to make sure the components/design works as expected. If you have that possibility I strongly recommend it.
Result:
Output voltage is according to spec. I measured 2.77-5.39V. Low max voltage can be caused by inacurate potentiometer, you should try to find one with as close as possible to 100kOhm.
The amount of current that can be loaded at different input voltage is automatically limited according to specification graph. When current increases, the voltage drops. If you need 500mA at 5V out for example, you need at least 4V input.
Next step: Construct the Circuit
Result:
Output voltage is according to spec. I measured 2.77-5.39V. Low max voltage can be caused by inacurate potentiometer, you should try to find one with as close as possible to 100kOhm.
The amount of current that can be loaded at different input voltage is automatically limited according to specification graph. When current increases, the voltage drops. If you need 500mA at 5V out for example, you need at least 4V input.
Next step: Construct the Circuit
Step 4: Construct the Circuit
Design:
I started to place all components as smart and space efficient as possible. The conductive copper lanes should be utilized as much as possible, priority two is to cut the lanes and priority three is to add external wires. To come up with an optimized layout with as few extra wires as possible is the most challenging part (maybe there are software that can do that). That could be avoided if a real circuit board is used or a smarter design. If you don´t want to make your own design, then just skip this step and copy mine. But I made it too tight at some places and it was a bit tricky to put wires in place. My recommendation is to place D2, C4 and everything above (in top view) one hole in up-direction. That results in more space for W2, W3 and W1 could be placed better. I later tried that modification and made another one, see picture.
Place Components (blue):
Place all components in correct place and fixate them with tape (to turn around the card without them falling out). See images to find correct spot for each component. Make sure electrolytic capacitors and diodes has correct polarization! The IC socket must also be correctly placed.
Soldering:
You should have some basic soldering experience. Prototype boards can be a bit tricky. Solder all components at once in this step. Double check that all components are in correct place and cut the wires as short as possible.
Cut (red):
Some lanes has to be cut in order to fulfill the layout. Use something thin and sharp and scratch away the copper at a width of about 0.5-1mm. See my image of where to cut. I used 10 cuts.
External Wires (green):
Some extra wires need to be added too, see picture. I had to add 7 extra wires. It´s safer to put them on the component side to not create shortcuts by mistake. W3 was very tight to place and had to be made a bit special (an old resistor-leg soldered to a cable). W7 was soldered on the copper side without insulation. Some cables need to be thicker in order to optimize efficiency in the circuit (I used cables from a computer ATX power supply):
I started to place all components as smart and space efficient as possible. The conductive copper lanes should be utilized as much as possible, priority two is to cut the lanes and priority three is to add external wires. To come up with an optimized layout with as few extra wires as possible is the most challenging part (maybe there are software that can do that). That could be avoided if a real circuit board is used or a smarter design. If you don´t want to make your own design, then just skip this step and copy mine. But I made it too tight at some places and it was a bit tricky to put wires in place. My recommendation is to place D2, C4 and everything above (in top view) one hole in up-direction. That results in more space for W2, W3 and W1 could be placed better. I later tried that modification and made another one, see picture.
Place Components (blue):
Place all components in correct place and fixate them with tape (to turn around the card without them falling out). See images to find correct spot for each component. Make sure electrolytic capacitors and diodes has correct polarization! The IC socket must also be correctly placed.
Soldering:
You should have some basic soldering experience. Prototype boards can be a bit tricky. Solder all components at once in this step. Double check that all components are in correct place and cut the wires as short as possible.
Cut (red):
Some lanes has to be cut in order to fulfill the layout. Use something thin and sharp and scratch away the copper at a width of about 0.5-1mm. See my image of where to cut. I used 10 cuts.
External Wires (green):
Some extra wires need to be added too, see picture. I had to add 7 extra wires. It´s safer to put them on the component side to not create shortcuts by mistake. W3 was very tight to place and had to be made a bit special (an old resistor-leg soldered to a cable). W7 was soldered on the copper side without insulation. Some cables need to be thicker in order to optimize efficiency in the circuit (I used cables from a computer ATX power supply):
- Between common ground (W1)
- Between pin8 (LX) and L1 (W2)
- Between pin6 (OUT) and D1 (W3)
Step 5: Testing and Finishing
Connect input power source and slowly increase the voltage (with unloaded output). The IC should start-up and LED should be lit at about 1V. If it has not started at 2V, then something is wrong, turn of power to not damage the IC!
I measured the output voltage to be as adjustable as before (2.77-5.39V). I measured efficiency at 2.5V/340mA in and 4.94V/150mA out. Converted to watts, (4.94*0.150)/(2.5*0.340)=0.74/0.85W, gives 87% efficiency. That is exactly according to spec but I got less efficiency at lower output load, but still expected. For full efficiency you need better components and a more compact layout. Some connections should not be more than 5mm according to PDF. There are two free copper lanes in my layout that can be utilized as common ground, I tried that too but got exactly the same result. It can also be my 10 year old multi-meter that is not calibrated.
I also measured the efficiency at maximum power. I had 3V/900mA in and 4.66V/460mA out. Converted to watts, (4.66*0.46)/(3*0.9)=2.14/2.7W, gives 79% efficiency.
I also successfully charged an iPhone 4s with it. It charge at 500mA with 3V in and 4.7V out. Decreasing the input voltage to 2V gives 4.4V out at 300mA (which is quite slow charging). The maximum charge for iPhone is 5V/1A from standard charger. This circuit cannot do that, see earlier suggestions for more suitable ICs.
I compared my construction with the Mintyboost version 3 kit from Adafruit. I charged my iPhone 4s and measured output effect with three different input voltages, 2, 3 and 4 volts.
MintyBoost:
2V in => 4.4V*300mA=1.3W out
3V in => 4.5V*400mA=1.8W out
4V in => 4.6V*470mA=2.2W out
My build:
2V in => 4.4V*300mA=1.3W out
3V in => 4.7V*500mA=2.35W out
4V in => 4.95V*500mA=2.5W out
They were similar at 2V but my construction is more powerful at higher voltage. These values might change when charging other phones, Apple products are quite strange to deal with..
Finishing:
As last step, cut the prototype board with a small saw. Two holes can be drilled at the corners if you want to mount it somewhere.
I measured the output voltage to be as adjustable as before (2.77-5.39V). I measured efficiency at 2.5V/340mA in and 4.94V/150mA out. Converted to watts, (4.94*0.150)/(2.5*0.340)=0.74/0.85W, gives 87% efficiency. That is exactly according to spec but I got less efficiency at lower output load, but still expected. For full efficiency you need better components and a more compact layout. Some connections should not be more than 5mm according to PDF. There are two free copper lanes in my layout that can be utilized as common ground, I tried that too but got exactly the same result. It can also be my 10 year old multi-meter that is not calibrated.
I also measured the efficiency at maximum power. I had 3V/900mA in and 4.66V/460mA out. Converted to watts, (4.66*0.46)/(3*0.9)=2.14/2.7W, gives 79% efficiency.
I also successfully charged an iPhone 4s with it. It charge at 500mA with 3V in and 4.7V out. Decreasing the input voltage to 2V gives 4.4V out at 300mA (which is quite slow charging). The maximum charge for iPhone is 5V/1A from standard charger. This circuit cannot do that, see earlier suggestions for more suitable ICs.
I compared my construction with the Mintyboost version 3 kit from Adafruit. I charged my iPhone 4s and measured output effect with three different input voltages, 2, 3 and 4 volts.
MintyBoost:
2V in => 4.4V*300mA=1.3W out
3V in => 4.5V*400mA=1.8W out
4V in => 4.6V*470mA=2.2W out
My build:
2V in => 4.4V*300mA=1.3W out
3V in => 4.7V*500mA=2.35W out
4V in => 4.95V*500mA=2.5W out
They were similar at 2V but my construction is more powerful at higher voltage. These values might change when charging other phones, Apple products are quite strange to deal with..
Finishing:
As last step, cut the prototype board with a small saw. Two holes can be drilled at the corners if you want to mount it somewhere.