AC to DC Converter 35V @5A err 1.5A

[Eshal]

Hello experts!

My task is to design 35V AC to DC converter at 5A. I will go simple all the way so I will use LM317 as a regulator.

** Changed to 1.5A

Stages of the design are given below:
1) Selecting transformer
2) Selecting diodes
3) Selecting input capacitor (value and working voltage) to the LM317
4) Selecting output capacitor (value and working voltage) to the LM317
5) Choosing resistor values

I need help from starting point. What considerations should I make in selecting transformer?

Thanks all.

 

[D.A.(Tony)Stewart]

Re: AC to DC Converter 35V @5A

Simple AC-DC is most expensive,most heat loss, biggest size and heaviest.

The transformer choices are limited to availability for 175W output @ 35Vdc. This may be useful for historical training of why 1970's style designs are no longer employed. Rather learn about all the SMPS designs from a good book such as author, teacher, designer Keith Billings with 40 yrs experience.

You left out the 200~240Vac/50Hz AC input >> or universal?
but you want AC to 35Vdc fixed @5A.

Because BJT type LDO's require >2.5V drop 5A is excessive heat lost which dumps 15W minimum and with input variation of AC and unregulated DC even more., So never specify that a design must use "xyz" without giving good reasons, its a bad start.

Because low RdsOn MOSFETS were NOT around in the 70's to be integrated in LDO's, big power supplies were in big racks. I used to use computer grade Capacitors that were the cylindrical 80mm x200mm or so with equally bulky XFMR's.


Core size can be reduced with rising F , switching transformer types are most popular,

 

[Eshal]

Re: AC to DC Converter 35V @5A

so you are asking me to design SMPS instead of the classical one with transformer? Right?

But I want to start from basic. I can't neglect basic just like for understanding Quantum we can't neglect Newtonian mechanics.

Any further suggestion would you like to give before going at the starting point? You are experienced as I read your profile so I can't move even a single step further without your help or the one like you.

- - - Updated - - -

so you are asking me to design SMPS instead of the classical one with transformer? Right?

But I want to start from basic. I can't neglect basic just like for understanding Quantum we can't neglect Newtonian mechanics.

Any further suggestion would you like to give before going at the starting point? You are experienced as I read your profile so I can't move even a single step further without your help or the one like you.

 

[ALERTLINKS]

Re: AC to DC Converter 35V @5A

LM317 outputs current in excess of 1.5A, but it is much less than 5A, as desired.
Here is its datasheet,
http://www.fairchildsemi.com/datasheets/LM/LM317.pdf

Here is a technique to increase output current,
**broken link removed**

 

[Eshal]

Re: AC to DC Converter 35V @5A

lol..
I know it is maximum useable for 1.5A. But I don't know what I was thinking that I wrote 5A. It is a mistake in my thread. I accept and thank you very much that you have mention it but unfortunately I can't edit my title.

 

[KlausST]

Re: AC to DC Converter 35V @5A

Hi,

There are a lot of "how to build a power supply" documents in the internet. Have you tried to calculate it on your own?

To design that simple sounding power supply one needs more information.
(do you only want to calculate it or build it in reality?)

* Input voltage and the min/max values of it.
* transformer type or at least no_load output voltage and full_load output voltage.
* do you want to spend more money for bigger capacitors to get a better efficiency / less heating?

Klaus

- - - Updated - - -

Hi,

not 5A...

then 1.5A? or 0.5A? or anything else?


Klaus

 

[D.A.(Tony)Stewart]

Re: AC to DC Converter 35V @5A

If you wish to browse other solutions, Google images works well for ideas.

Use keywords like schematic too.

 

[Eshal]

Re: AC to DC Converter 35V @5A

There are a lot of "how to build a power supply" documents in the internet. Have you tried to calculate it on your own?

I know there are many articles but I want my own. Yes I tried but stuck everywhere during design.

To design that simple sounding power supply one needs more information.
(do you only want to calculate it or build it in reality?)

Only calculation.

* Input voltage and the min/max values of it.
* transformer type or at least no_load output voltage and full_load output voltage.
* do you want to spend more money for bigger capacitors to get a better efficiency / less heating?

* should operate at 230AC/220AC
* simple transformer not center tapped and at least no load voltage should be 35V but i don't know how to choose full load voltages.
* if I am going to make it in reality then yes i am able to spend more money for bigger capacitors but i am going to only design it with knowledge and formulas and theory

If you wish to browse other solutions, Google images works well for ideas.

Use keywords like schematic too.

I want my own.

 

[THINKER_KAM_MAN]

Re: AC to DC Converter 35V @5A

just a question

have you got a transformer? is it 35V@5A, 200VA transformer.
torrid?. it be ok.

you just need bridge rectifier, smoothing capacitor, 723 regulator
driver transistor (TIPxxx), series pass transistor (5 off 2n3055 in parallel)
with 0.22R current shearing resistors in the emitters
and it be simple. ho, also heat sink and if you want a fan.
you be connected to the mains so why bother you self with a switcher.
for efficiency, will it be cheaper? will it be more reliable, will it be less noisy, will it be easy to get it to work??? but if you want to switch
there are many 3 terminal offline power regulators which could be used
they will require input filter, isolating transformers, high voltage capacitors and diodes and bridge, optical isolater, special low esr caps
low voltage drop diodes, good circuit layout. and you may have to
wind you own transformer. well yes that's all a lot to do.
advantage less weight, save power, but there is lots of power in the mains when at home. (ho may be its a green thing)
also of course you could use a mains transformer and a 3 terminal
SMP. but you will need a Choke, and all the other stuff baring the
HV cap.
so i agree with you use a mains transformer not efficient
as SMP but reliable and easy to design and get to work

 

[D.A.(Tony)Stewart]

Re: AC to DC Converter 35V @5A

Thank Keith Billings for this AC RMS to DC chart vs W based on an RC filter with 1.5 uF/W for 60Hz and 1.8uF/W for 50Hz.
thus 35V@5A=175W, C=315uF is the minimum for this chart to be valid.

In his book, he has similar nomographs for ripple Cap selection, but Using low pass filter theory, choose T=1/f for desired ripple of say 10% of 100Hz. we know a 1st order filter is -20dB/decade so 10%= -20dB or a breakpoint of 10Hz. For a 175W supply, 5% -10% loss in the filter resistor is reasonable which reduces the diode/ capacitor peak charge current or ripple current vs R loss as a tradeoff. Less ripple current means more reliability and less heat in the capacitor for I^2ESR loss.. , so 5% is 8W and 10% is 17.5W.

Using P=8W and max Idc=5A ,

Rs=I²dc/P= 3.1 so from the chart below , Interpolating at 175W and Rs=3.1, Vdc (mean)/Vrms (min) at rated power = 1.30

LDO drop can be chosen by design from near 0 to 3V depending on choices of series switch. A 10 mOhm MOSFET only drops 50mV at 5A while old LDO's like LM338 adjustable 5A regulators are <2.5V at 25'C and rises at extreme temps. But the LDO integrated deign offers thermal protection and adjustable voltage or adj. current limit. But it's CC limit design requires 2.5 + 1.25Vref drop as well, which is a power loss cost penalty if you choose this.

For adjustable CV only, Vdc unregulated is now 35V+2.5=37.5Vmin with 1.30 ratio and Vac=37.5/1.3=29V rounded up.
Choose your input supply range minimum and thus you have your transformer voltage ratio with the power rated in VA's being 175+8W + Bridge loss of 2V @5A or 10W ending up with a 200VA transformer required. The bridge can be Schottky 60V@ 10A for lower temperature rise that should still be mounted on a heat sink.

RC values for LPF

recall Rs=3.1 @8W or closest value ceramic resistor.
also 10% ripple lead to f=10Hz or C=1/(2πfRs)= 5 mF @ 1.5x unregulated no load voltage for safety margin and reliability. No load voltage will be only Vdcmax=Vdc mean + Vripple plus Line variance for overvoltage and transients. So 150% of 40V is an expensive cap for 10% ripple namely 5mF @60V https://www.digikey.com/product-detail/en/CGO502M055L/CGO502M055L-ND/1595674

Now that you can see that this cap is unsuitable $ for this low power , choose 20% ripple and redo the math and use a liitle more than 5% loss in Rs. https://www.digikey.com/product-detail/en/B12J3R0E/B12J3R0E-ND/823477

Design can be an interactive process if you haven't done this for a while.

Consider a Microwave Oven transformer in reverse or rewind your own secondary to get a custom full load voltage of x Watts of Y Vrms

 

[D.A.(Tony)Stewart]

Re: AC to DC Converter 35V @5A

After simulating my quick design. I realized my formula for RC filter was based on linear systems, which this is not for filtering..
It appeared to me that the ideal Rs value was around 10% of the effective load resistance of 175W/35V=7 OHms so Rs is 0.7 Ohms.

Second I found cheaper capacitors 6800uF @ 100V @$8,23 (6.8mF) which improves the lower level voltage on ripple. Lowering Rs only increases the peak ripple voltage without affecting the Vmin peak . I ended up with 23% ripple pp 36~42Vdc with 7 Ohm load with 11Ap current on 5Adc. by doubling up on the capacitor.

LDO not shown

You can play with my SIM and end up with your own design and add transformer ESR to see how it affects the result.
Capacitor ESR is no longer important with the Rs filter.
The transformer turns ratio selected for 230Vac RMS was 6.25:1 voltage or 36.8Vrms to get 39Vdc avg out at load for a ratio of 1.06 which means peak voltage of 55V which it reaches with no load meaning a high ratio and inherent inefficiency of 50 Hz linear designs.

 

[KlausST]

Re: AC to DC Converter 35V @5A

Hi,

I want my own.

What´s the difference between googling the internet for documents and asking us?

If you want your own, then tell us your ideas, specifications and show us schematics.

For sure we can design a power supply for you..


Klaus

 

[Eshal]

Re: AC to DC Converter 35V @5A

Here is the schematic.



Want to know what should be values of the components in it.

Thanks all for great helps.

 

[D.A.(Tony)Stewart]

Re: AC to DC Converter 35V @5A

I want my own.

To help you learn, you must first try yourself then ask for help to check. Be sure to include a tolerance on AC inputs. You have used up my limit of hand-holding. Consider RC bridge filter or LCR bridge filter.

 

[Audioguru]

Re: AC to DC Converter 35V @5A

Eshal,
It looks like you almost copied the circuit shown in the datasheet for the LM317. Do you need a variable output voltage?
The calculation for the resistors is shown in the datasheet. HINT: The circuits all show the more expensive LM117 that can use 240 ohms for R1. The cheaper LM317 needs double the resistor current so its R1 should be 120 ohms, then R2 should also have half the resistance shown.

 

[Eshal]

Re: AC to DC Converter 35V @5A

@Audioguru

It looks like you almost copied the circuit shown in the datasheet for the LM317.

No no. It is not copied. I draw it on multisim with the help of my BRAIN.

Do you need a variable output voltage?

Yes variable output voltage, output voltage should be maximum 30V to 35V with output current 1A to 1.5A

The calculation for the resistors is shown in the datasheet.

Which data sheet? LM317?

The circuits all show the more expensive LM117 that can use 240 ohms for R1. The cheaper LM317 needs double the resistor current so its R1 should be 120 ohms

What does it mean dear? I means for LM117 R1=240 and for LM317 R1=120. Why is it so?

then R2 should also have half the resistance shown.

I know about this rule.

@SunnySkyguy

To help you learn, you must first try yourself then ask for help to check. Be sure to include a tolerance on AC inputs. You have used up my limit of hand-holding. Consider RC bridge filter or LCR bridge filter.

I tried to calculate values of the components in the given figure but I am stuck many places. I tried by self but obviously I am only alone what to do if I stuck, ofcourse I will move toward the edaboard.com and for filteration you are asking me to use LCR?

- - - Updated - - -

@Audioguru
Here is the datasheet of LM317. Tell me which circuit I have copied? You will not find any circuit same as the one in post#13 because I have not copied it from the datasheet even I draw it with my mind with knowledge what components should I use. And also tell me where is the calculation of resistor in the datasheet too? Basic formulas are given with unknown parameters. Until I don't choose those parameters unless I can't calculate the resistor value.

 

[KlausST]

Re: AC to DC Converter 35V @5A

Hi

No no. It is not copied. I draw it on multisim with the help of my BRAIN.

It´s the usual way to take the schematics of the datasheets and use it for one own design. Often one has to modify something to meet the desired function.
In the datasheets there are proven circuits (hopefully ;-) ) with hints and formulas to calculate the device values.

Basic formulas are given with unknown parameters.

The most important formula in datasheet is: Vout = 1.25V x (1 + R2/R1) + Iadj x R2

Vout = known by you (Eshal)
R1 = 120 or 240 depending on regualtor
Iadj = given in datasheet.
--> R2 is the only one unknown value.

So you can solve the formula for R2.

This is the same way we all have to go.

*********
I just googled for "lm317 power supply"

There are a lot of good results:
* youtube videos on how to design your own supply
* online tutorials on how to design
* schematics (you may use ist. no need to reinvent the wheel)
* PDF files with standard designs
* pdf files on how to extend on voltage and/or current.

also:
There are a lot of manufacturers for LM317. They usually have additional application notes on how to use the regulators. Go through them.

*********

You are always welcomo to ask your questions here, but please be more specific. (on where you stuck, or what parameters are unknown)

Klaus

 

[BradtheRad]

Re: AC to DC Converter 35V @5A

Want to know what should be values of the components in it.

I have been planning to run a Falstad simulation and post a diagram, but SunnySkyguy did the honors already (post #11).

I suggest you click his link 'my SIM'.

Falstad's simulator is animated and interactive. You can watch electrons (or rather, current bundles) move through wires, at a speed which is proportional to amperes. Hover the mouse over a component to watch its specs moment-to-moment.

Observe the value of current bursts going through diodes.

Hover over a scope trace, to turn its component blue.

The load is switched. Observe variations in output voltage.

Observe how much amperage goes back and forth through the smoothing capacitor.

Although Falstad's does not contain a 317 or 117 regulator IC, it is easy to use, and to experiment with.

 

[schmitt trigger]

Re: AC to DC Converter 35V @5A

If you require 5 amps, better use the LM338. The LM317 is only useable to 1.5 amp.

Both operate identical, so your drawing is good...you only have to upgrade the other components to handle the additional current.

 

[Audioguru]

Re: AC to DC Converter 35V @5A

@Audioguru

No no. It is not copied. I draw it on multisim with the help of my BRAIN.

See figure 24 in your datasheet. But it uses 240 ohms for R1 because it uses an LM117 instead of an LM317.

Yes variable output voltage, output voltage should be maximum 30V to 35V with output current 1A to 1.5A

The figure 24 in the datasheet shows a graph of typical dropout voltage that is 2.25V with an output at 1.5A. Then for an output of 35V its minimum input should be 37.25V. Maybe your input is 40V and the output is 30V so the heating will be (40V - 30V) x 1.5A= 15W which will require a pretty big heatsink.

What does it mean dear? I means for LM117 R1=240 and for LM317 R1=120. Why is it so?

The datasheet explains that all the operating current of the regulator goes to the output, not to the ADJ terminal. Then the printed spec's show the Minimum Load Current is 5mA for an LM117 but is 10mA for an LM317. The datasheet says that without a load the output voltage will rise if there is not enough load current provided by the resistors. R1 has a minimum of 1.2V across it and an LM317 has a Minimum Load Current of 10mA so the maximum value of R1 should be 1.2V/10mA= 120 ohms unless the circuit has another load resistor.

Instead of having R2 parallel to your voltage adjustment variable resistor you should have a resistor in series with the variable resistor to limit its adjustment range to a minimum of 30V.