Jump to content
Electronics-Lab.com Community

varible voltage, current limiting, power supply


Recommended Posts

Im currently designing (slowly as i have little free time) a varible voltage current limiting power supply, below are some of the features that i would like to incorporate;

(bare with me as i ramble and am in the design/thinking stage still)

varible rails
0 to +30v (Imax ~2A)
0 to -30v (Imax ~1.5A)

constant rail
5v Imax ~2A

current limiting
have no idea quite what todo but want to build this feature in

AD 8051 (i got a couple of free samples, 14bit adc, 7 channels)
lcd measuring voltage and current.

Ok so i have read through google and this site and have got some really good ideas ;) The reason i have a constant 5v line is beuase i use 5v loads and want to be able to use other voltages at the same time.

Ive been looking at many ways to do things and pretty sure on most things except for the current aspect of the design. If i use the LM317 and LM337 (for the varible rate supply) (or the bigger brothers of these ic's capable of 3A and 5A) can i somehow incorporate current limiting?

+i trying to think of a way to measure current and am abit lost, ill probably use an op amp for voltage samples and then use the microcontrollers adc (5v input scale).

+if i want a constant rail as well as +/- 0 to 30v it means needing 2 transformers or using a 5v regulator although this isnt very efficient (burning off 25v) so should i simply use 2 transformers and 2 bridge rectifiers of pull everything off 1 transformer (they costs a fortune over here so id rather use 1)

thanks for any advice on what not to do or what i could do/use


Link to comment
Share on other sites

This power supply is exactly what i was thinking about. But it's difficult for me to desing it also :-\. Let's wait a more experienced user to drive us to the right direction!

Meantime, i suggest you to read:

0-30 Vdc Stabilized Power Supply With Current Control 0.002-3 A

which has current limiter circuit so you can take some ideas.

Also i have to make another "feature" proposion:

You can replace classic pontesiometers with DS1869 digital pontesiometers and have up down push buttons for adjusting the voltage.

Link to comment
Share on other sites

thanks ill have a look, ill draw up al my ideas into a design and keep all my developments to this thread ;)

i have had a look through the articles on this site and here aswell http://www.discovercircuits.com/P/pwr-variable.htm (moderators please delete if in-appropriate).

ive started to get an idea of the whole parts list and chassis dimensions.

If you want to use a micro-controller to display the voltage and current have a look at analogue devices ADUC812, an excellant microcontroller for applications like this, although it is abit excessive and

Link to comment
Share on other sites

I am also building a power supply with current limiting. For the current limiting feature I'll be using an SG-3524 Pulse width modulator and a low ohmic current sense resistor, 0.05 Ohms.

The basic theory is: Voltage drop across a low ohmic resistor is directly proportional to the current flowing through it. The voltage drop across the 0.05 Ohm resistor is fed into a LM324 OP-AMP, which amplifies the voltage drop by a factor of 10. This is then passed onto the PWM and is compared against a reference voltage which is set by a potentionometer. This is what allows you to set the current limit. Once the two voltages have been compared the output pulse from the PWM is either increased (more current) or decreased (less current) to maintain the current limit set by the pot. The output pulse train then drives an N-Channel Power MosFET to switch the higher current supply. For my application this switching will occour 20,000 times a second, (20Khz).

For example, if you set the current limit to 2A, the voltage drop across the resistor would be I*I*R


So that's a drop of 0.2V the LM324 then multiplies this by 10 to give 2V. That 2V is then compared against a reference voltage in the PWM which can be set to anywhere between 0V and 5V using the pot. As we have theoretically set the limit a 2A, the pot will be set to supply a 2V reference voltage. So we have 2V Ref and 2V from the LM324. These two voltages are equal so no action is taken to change the pulse width and 2A continues to flow.

Now, if we were running a motor from this supply and it suddenly stalls, trying to pull 3A. The voltage drop is now 0.45V*10=4.5V.

This 4.5V is now compared to our 2V Ref and the PWM has to take action to bring the 4.5V back to 2V. It does this by narrowing the width of the pulse to reduce the current flowing through the 0.05 Ohm resistor. The pulse width is narrowed until 2A and the resulting 0.2V drop *10=2V is achieved.

These parts are inexpensive. The PWM chip and a low Ohmic resistor can be had for around

Link to comment
Share on other sites

Hi, MylarHornet

ok its late so im going to go through what you said, in theory it sounds good and at a quick look its cheaper then this example (in the 'on' semicondutor datasheet)(thought it might be useful)

i just spent some time looking through retail psu's to get an idea of comman specs so i can design mine with the best usable features. i might include two adjustable postive rails and a single negative rail. Ive come to this conclusion after thinking about how iam going to be using it and certain situations back at uni where i had to use 2 psu's

ive hunted down the best price for spec transformer (well they all cost a fortune in the uk), 25v (DC 35.25V), 500VA/10A max(i figure if i have 2x1.5A max and 1x3A +other bits 10 should cover it), toroidal unit. Now im pretty much straight out of uni so theory wise it sounds ok and is also the cheapest transformer of its kind but is toroidal ok for this purpose ?

Anyway its by no means bought yet so if its wrong no problem ;)


Link to comment
Share on other sites

Hi, yes I'm in the UK too and you're right about prices. I'm shopping at the places you mentioned also. Sometimes bargains can be found in the advertisers in EPE Magazine.

I'm planning on using an open frame toroidal Xformer for my PSU. Should be fine for you also.

My design is for running variable speed DC motors at constant torque. So my current limiter is really a constant current source.

If I have a motor rated at 12V 1A, I can supply it with any voltage over 12V and the current delivered will always be 1A. I could give it 100V and it would still only draw 1A. Without the PWM the motor would be trying to pull 8.33A at 100V and it would be dead very quickly.

I'm not an electronics expert, but I've done alot of reading on the subject in relation to my particular application. Although you may find this approach useable, you may be better served with a design similar to the schematic you posted.

My design does not require variable voltage rails. You could for example run an LED, directly connected to 40V as long as you set the current limit to the current draw of the LED, say 10ma, the pulse width would be modulated to supply just the right amount of power to the LED, without blowing it to bits.

Pulse width modulation relies on having a higher voltage supply than that required to run the load.

Regards MH...

Link to comment
Share on other sites

  • 3 weeks later...

well, instead of using the LCD and PIC for dispaying the voltage and current, don't you think you could use a voltmeter and ampermeter kit wich display on 7-segments leds?
also, another suggestion. for the negative you could use a LM317 regulator, as it can provide 1.5 amps. i am not sure how it will cope with low voltage and high current since you have a higi input voltage, more than 30 volts, than you will have a lot of power disipated on the reg. might use 2 in series as suggested.

Link to comment
Share on other sites

  • 3 months later...

You have selected a challenging project, designing a power supply.
Many details must be observed, and perhaps you have overlooked this one: The LM317 and its higher current cousins will current-limit when the input-output voltage exceeds 15V. This is for their Safe Operating Area in order to prevent secondary breakdown. With an input voltage of 35V then current-limiting will occur if the output voltage is 20V or less and at a 5V output it will give only a few hundred mA.
A solution to this problem would be a paralleled booster transistor (PNP for low voltage drop) as shown on the spec sheets.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


  • Create New...