Global Sources
EE Times-India
Stay in touch with EE Times India
 
EE Times-India > Amplifiers/Converters
 
 
Amplifiers/Converters  

Step-Up DC-DC Converter Calibration and Adjustment Using a Digital Potentiometer

Posted: 28 Jun 2003     Print Version  Bookmark and Share

Keywords:power 

/ARTICLES/2003JUN/A/2003JUN28_AMD_POW_AN5.PDF

1 of 8 010903

Introduction

The purpose of this application note is to show an example of how a digital potentiometer can be used in the

feedback loop of a step-up DC-DC converter to provide calibration and/or adjustment of the output voltage.

The example circuit uses a MAX5025 step-up DC-DC converter (capable of generating up to 36V, 120mW

max) in conjunction with a DS1845, 256 position, NV digital potentiometer. For this example, the desired

output voltage is 32V, which is generated from an input supply of 5V. The output voltage can be adjusted in

35mV increments (near 32V) and span a range wide enough to account for resistance, potentiometer and DC-

DC converter tolerances (27.6V to 36.7V).

While the intent of this application note is to show an example of a step-up DC-DC converter, the ideas

presented here can be applied to generate other combinations of output voltages, step sizes, ranges, and power

requirements to meet your particular applications' needs. An additional application note is available (AN225)

that shows an example of using a digital potentiometer with a step-down DC-DC converter. A link to AN225

appears at the end of this document.

Fixed Step-up DC-DC Converter

The typical circuit from the MAX5025 data sheet is shown in Figure 1. In this circuit, the output voltage,

VOUT, is determined by the ratio of fixed resistors R1 and R2. These two resistors form a voltage divider that

feeds a fraction of the output voltage back to the FB pin, creating a closed-loop system. The system is at

equilibrium when VOUT is generating the desired output voltage and the R1 and R2 voltage divider feeds back

1.25V to the FB pin. When VOUT is lower than the desired output voltage (and hence the voltage fed back to

FB is below 1.25V), the DC-DC converter IC attempts to deliver additional power until FB reaches 1.25V.

Equation 1 is directly from the MAX5025 data sheet. Solving Equation 1 for VOUT yields Equation 2 where

VREF, the FB Set Point, is 1.25V for the MAX5025.

ww

x

v

gg

h

f

-= 1

V

V

R2R1

REF

OUT

Equation 1

w

x

v

g

h

f

+= 1

R2

R1

VV REFOUT Equation 2

Application Note 226

Step-Up DC-DC Converter Calibration and

Adjustment Using a Digital Potentiometer

www.maxim-ic.com

AN226

2 of 8

Figure 1. Fixed Step-Up DC-DC Converter Circuit

Digital Potentiometer Considerations

One possible way of adding a digital potentiometer into the feedback loop is shown in Figure 2. However,

before selecting a digital potentiometer, a number of considerations must be given some thought. The

following list of questions address important design considerations and assist in the device selection process.

1. Is a 3V or 5V supply available for the potentiometer?

The majority of digital potentiometers available require 3V or 5V to operate. Likewise, the voltage needs

to be available before VOUT can be generated. Since the example circuit has a 5V input supply, this is not

a concern here but if VIN were larger, this would be a major concern.

2. Will the system only be calibrated once? Or monitored/controlled closed loop? How will the digital

potentiometer be controlled? Using a microcontroller? Push-button?

The answer to these questions will help select a specific digital potentiometer. For example, if the system

is going to be calibrated once, say during production testing, then a NV (NV) potentiometer is needed to

save the calibrated wiper position. Likewise, it is also important to consider how the potentiometer will be

programmed. Will it be programmed with a production tester capable of talking 2-wire or 3-wire, or is a

push-button potentiometer needed?

3. How many steps are needed? What resolution is required? What range is needed?

These answers determine the minimum acceptable number of potentiometer positions and serve as a

benchmark when experimenting with different resistor values when finding a combination that is

acceptable even in worst case conditions due to resistor/component tolerances. Furthermore, be aware

that the desired output voltage range will be different than the actual output range so that the desired

range can also be obtained even when using worst-case tolerances.

4. What is the max voltage that will ever be applied to the potentiometer terminals? Is VREF (VFB) larger than

the maximum allowable voltage on the VW pin?

Some potentiometers have a specification that states the maximum allowable voltage than can be applied

to any of the potentiometer terminals. The second question is intended to be a sanity check. For example,

the MAX5027 (not the MAX5025) has a VREF of 30V! This voltage cannot be applied to the wiper

terminal. The reason VREF is so high is because it is a non-adjustable DC-DC converter.

MAX5025

PGND

/SHDN

VCC

LX

FB

GND

U1

VCC

4.7uF

C1

C2R1L1

D1

C3*

R3*

R2

47uH

ZETEX SCHOTTKY DIODE

VOUT

+

-

100W

1uF, 50V1uF, 50V

UNFILTERED

VOUT

*Optional RC filter

1%

1%

VCC = VIN = 5V

VIN

= 5V

+

-

(fixed)

AN226

3 of 8

5. Is the FB pin bias current large enough to cause a noticeable drop across the wiper resistance, RW? Does

the FB pin bias current exceed the maximum (or abs max) wiper current specification?

Fortunately, the maximum wiper current specification is usually 1mA or larger, while the bias current is

in the magnitude of nA. The second reason why the bias current is important is because if too large, the

voltage drop across the wiper resistance becomes noticeable and should be accounted for when

calculating the output voltage.

6. Does the minimum and maximum position of the potentiometer connect directly to VL and VH?

For example, the DS1805 is different in that the maximum potentiometer setting (255) does not connect

directly to the H terminal. It is actually one resistor (LSB) away from the H terminal. Granted, the

difference between 255 and 256 resistors are nil, the same is not the case for potentiometers with fewer

positions. The result of this question can be seen later in Equation 4.

7. What is the desired tolerance of VOUT? 5%? 10%? 3%?

Keep in mind that the DC-DC converter IC itself has a tolerance. The MAX5025 for example has a

tolerance of 5%. And this is completely independent from the tolerance of R1 and R2, which may also be

5% or 1%. These tolerances will be discussed in detail later.

8. What is the operating temperature range of the circuit?

Just as important as knowing the effects of the tolerances, it is equally important to know the effects of

temperature on the output voltage.

Figure 2. Digital Potentiometer in the Voltage Divider Feedback Path

Adding a Digital Potentiometer to the Feedback Path

Although there are several ways that a digital potentiometer could have been added to the circuit in Figure 1,

this application note will only discuss the voltage divider configuration as shown in Figure 2. Two

configurations that will not be discussed are 1) using the pot as a variable resistor (by connecting the wiper

terminal to either the high or low terminal) between R1 and R2, and 2) eliminating R2 and connecting the pot

low terminal to ground.

R1

R2

Adjustable VOUT

RH

RL

VH

VL

VW

RW(To FB pin of

DC-DC Converter)

VFB

= 1.25V

IW

Caution must be taken so

that VH does not exceed the

specification of the digital

potentiometer.

IW due to MAX5025

is 110nA typical.

RPOT

RPOT

= RH

+ RL

RL

= position x (RPOT

/(total # of positions - 1))

RH = RPOT - RL

AN226

4 of 8

Equations 3, 4, and 5 show the relation of the potentiometers' position to RL, RH, and RPOT. Once a

potentiometer is added into the feedback loop, these equations can be used to modify Equations 1 and 2 to

represent the new circuit.

LHPOT RRR += Equation 3

ww

x

v

gg

h

f

-

4=

)1(

R

R POT

L

positionspotofnumbertotal

positionterpotentiomecurrent Equation 4

LPOTH R-RR = Equation 5

RPOT is the end-to-end resistance of the potentiometer and RH and RL are dependent on the current wiper

position setting (see Figure 2). The denominator of Equation 4 is (total number of pot positions -1) if the max

and min wiper positions connect directly to the H and L potentiometer terminals, which is true for the

DS1845. Some digital potentiometers may have an additional resistor between the max potentiometer setting

and the H terminal. The denominator for those potentiometers would not include the -1 and simply be (total

number of pot positions).

Equation 2 becomes,

ww

x

v

gg

h

f

+

+

+

= 1

RR2

RR1

VV

L

H

REFOUT Equation 6

Resistor Calculations

Unfortunately, there is no quick, easy way of calculating R1 and R2. VOUT is a function of multiple variables

and many solutions exist. Choosing the optimal solution for a particular application involves a decent amount

of trial and error. For this reason, a spreadsheet is the single most valuable tool because it allows the designer

to make a tweak and instantly see its effects on VOUT for a single potentiometer position as well as a sweep of

the entire potentiometer range. The following will describe the process used to calculate the values for the

example circuit generating 32V.

Assumptions made up front for this design were, RPOT = 10kW, 256 positions, VREF = VFB = 1.25V, and the

output voltage will be 32V when the pot is set in the middle position (position 127). The spreadsheet created

for this application note can be found on the Dallas Semiconductor ftp site. A link to the spreadsheet can be

found at the end of this application note. Plugging the assumptions into Equations 4 and 5, and then into

Equation 6 determines the needed ratio between R1 and R2.

6.241

25.1

32

4.9804R2

6.0195R1

1-

V

V

REF

OUT

=-=w

x

v

g

h

f

+

+

= Equation 7

The MAX5025 data sheet says to choose an R2 in the 5kW to 50kW range and then calculate R1. However,

how can one make an educated R2 selection in such a large range without knowing what else will be

effected? The spreadsheet shown in Figure 3 was created specifically for this. Simply type in a value for R2

in the red cell (D7). R1 is then automatically calculated to obtain the ratio in cell Q7 (which was the result of

Equation 7). In addition, thousands of other calculations are also performed and then VOUT vs. Pot Position is

plotted. The VOUT vs. Pot Position plot shows the expected output voltage for all 256 potentiometer positions.

Linearity, range, and slope can all be seen clearly. Look for the trace labeled "Typical". The other traces will

be described later. Comparing plots of R2 = 5kW and R2 = 50kW (Figure 4) it can be seen that smaller values

of R2 (Figure 4a) produce a much larger output range, although non-linear and having a steep slope. The

steep slope produces larger than desired step sizes (especially at the lower potentiometer positions)

AN226

5 of 8

decreasing the resolution of VOUT. Larger values of R2, on the other hand, produce a linear output and much

smaller slope (Figure 4b). The smaller the slope, the smaller the step size (and hence resolution). When

attempting to "fine-tune" a particular output voltage, it is desirable to have many potentiometer positions at

and around the desired output voltage. The drawback, which will be discussed in the following section, is that

since R1 and R2 have a tolerance of 11%, there is a possibility that no potentiometer setting will reach 32V.

This can be seen looking at the top trace in Figure 4b. When the potentiometer is set to position 0, the output

voltage is ~37.5V and when the potentiometer is set to position 255, the output only goes down to 32.3V.

Therefore, it is important to find a happy medium. For the example circuit, the happy medium (determined by

trial and error, entering various values of R2 and looking at the VOUT vs. Pot Position graph ensuring that 32V

could be reached in all conditions with an acceptable step size) is 30kW. The closest 1% SMT standard value

is 30.1kW. Plugging this value into the spreadsheet and having it recalculate R1, the closest standard value

had to be found for R1 as well. The graph shown in Figure 3 shows that 32V can be obtained even in the

worst-case conditions.

Figure 3. Example Resistor and Error Spreadsheet Screenshot

AN226

6 of 8

Once values for R1 and R2 are selected, it is important to verify that VH does not exceed the maximum

specification (of VCC + 0.5V for the DS1845). VH can be calculated using Equation 8.

( )

( )POT

POT

OUT

H RR2

RR2R1

V

V +4

++

= Equation 8

Figure 5 shows a screenshot of the bottom of the example spreadsheet. The calculation of VH can be seen on

the far right along with the max voltage across the potentiometer. We can see that for the resistor values we

chose VH has a worst case potential of 1.84V. This voltage is well within the recommended operating

conditions. Other items in interest in Figure 5 are VOUT min/max, min/max deltas between pot positions, and

also min/max VH. While some of the data and calculations shown in the spreadsheet may appear to be of little

use, it provides multiple sanity checks to ensure a good design.

Figure 4. R2 Comparison

Figure 5. Additional Calculations

VOUT vs. POT POSITION when R2 = 5kW

25

27

29

31

33

35

37

39

41

43

45

47

49

51

53

55

57

59

61

63

65

67

69

71

0 50 100 150 200 250

POT POSITION (DEC)

VOUT(V)

Typical

Ratio Min/ Pot Min

Ratio Min/Pot Max

Ratio Max/Pot Min

Ratio Max/Pot Max

R1 = 240k (1%),

R2 = 5k (1%),

POT = 10k (20%),

VFB = 1.25V (5%)

Typical

VOUT vs. POT POSITION when R2 = 50kW

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

0 50 100 150 200 250

POT POSITION (DEC)

VOUT(V)

Typical

Ratio Min/ Pot Min

Ratio Min/Pot Max

Ratio Max/Pot Min

Ratio Max/Pot Max

R1 = 1.3M (1%),

R2 = 50k (1%),

POT = 10k (20%),

VFB

= 1.25V (5%)

Typical

a) R2 = 5kW b) R2 = 50kW

AN226

7 of 8

Error Analysis

Up to this point, all calculations have used typical (nominal) values. However, to ensure the design is

production worthy it is essential to calculate the variations of the output voltage due to component tolerances,

temperature variations, and any other sources of error and ensure that the desired output voltage can always

be obtained. This application note will go as far as analyzing the effects of resistor, potentiometer, and VREF

tolerances. The analysis of temperature variations, however, is saved for a future application note.

The MAX5025 VREF has a tolerance of 15%. This means that the 1.25V could actually be anywhere between

1.19V and 1.31V. What makes this tolerance different than that of the resistors and potentiometer is that this

5% is for the entire temperature range. The resistors and potentiometer on the other hand spec both a

tolerance and a temperature coefficient. Resistors R1 and R2 both have a tolerance of 11%. The tolerance of

the DS1845 is 120%.

Referring back to Figure 3, the use of these tolerances can be seen in row 7 of the spreadsheet. For example,

C7 is the nominal value of R1, while J7 is nominal (C7) minus 1% and K7 is the nominal plus 1%. Once this

is done for all of the tolerances, calculations can easily be repeated multiple times, calculating all possible

combinations in search of the combinations that yield the minimum and maximum output voltages. These

combinations can then be added to the VOUT vs. Pot Position plot and verified that each can generate 32V.

Once all of the traces on VOUT vs. Pot Position graph meets the desired specifications, the resistor selection is

complete. Figure 6 shows the final circuit with the DS1845 and with the selected values of R1 and R2. The

VOUT vs. Pot Position plot for this circuit is shown in Figure 3.

Figure 6. Final Circuit Using a DS1845 Digital Potentiometer

MAX5025

PGND

/SHDN

VCC

LX

FB

GND

U1

VCC

4.7uF

C1

C2R1L1

D1

C3*

R3*

R2

47uH

ZETEX SCHOTTKY DIODE

VOUT = 32V

+

-

100W

845kW

30.1kW

1uF, 50V

1uF, 50V

UNFILTERED

VOUT

= 32V

*Optional RC filter

Address=000

DS1845

W1

L0

W0

SDA

SCL H0

VCC

H1

U2

A0

A1

A2

WP

GND

L1

VCC

VCC

4.7kW

R4 R5

4.7kW

To 2-Wire

Interface .1uFC4

VCC

1%

1%

VCC = VIN = 5V

VIN

= 5V

+

-

(adjustable

from 27.6V to

36.7V in 35mV

increments,

typical)

AN226

8 of 8

Conclusion

This application note shows an example of how to use a digital potentiometer in the feedback loop of a step-

up DC-DC converter to allow the output voltage to be calibrated. While this application note specifically uses

the MAX5025 and the DS1845 to generate 32V, the concepts presented here can be applied towards other

potentiometer/converter combinations as well as other output voltages and power ratings.

Questions/comments/suggestions concerning this application note can be sent to:

MixedSignal.Apps@dalsemi.com.

Link to the spreadsheet used in this example:

ftp.dalsemi.com/pub/system_extension/pots/AN226/AN226.xls

Link to Application Note 225 showing a step-down DC-DC converter:

http://pdfserv.maxim-ic.com/arpdf/AppNotes/app225.pdf

Maxim Integrated Products / Dallas Semiconductor Contact Information

Dallas Semiconductor

4401 S. Beltwood Parkway

Dallas, TX 75244

Tel: 972-371-4448

Maxim Integrated Products, Inc

120 San Gabriel Drive

Sunnyvale, CA 94086

Tel: 408-737-7600

Product Literature / Samples Requests:

(800) 998-8800

Sales and Customer Service:

(408) 737-7600

World Wide Website:

www.maxim-ic.com

Product Information:

http://www.maxim-ic.com/MaximProducts/products.htm

Ordering Information:

http://www.maxim-ic.com/BuyMaxim/Sales.htm

FTP Site:

ftp://ftp.dalsemi.com

Tech Support:

MixedSignal.Apps@dalsemi.com





Comment on "Step-Up DC-DC Converter Calibration ..."
Comments:  
*  You can enter [0] more charecters.
*Verify code:
 
 
Webinars

Seminars

Visit Asia Webinars to learn about the latest in technology and get practical design tips.

 

Go to top             Connect on Facebook      Follow us on Twitter      Follow us on Orkut

 
Back to Top