Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Not being able to understand simple LM317 lead acid battery charger

Status
Not open for further replies.
Rather than SIMULATING these Circuits, Why don't you just BUILD THEM?

Just using the LM317, It Will EASILY Charge up a 1,000uF Capacitor, or even a 10,000uF Cap.

Why the Additional Transistor?
How Much Current do you need.
 
Last edited:
Rather than SIMULATING these Circuits, Why don't you just BUILD THEM?

Just using the LM317, It Will EASILY Charge up a 1,000uF Capacitor, or even a 10,000uF Cap.

Why the Additional Transistor?
How Much Current do you need.
I do not have oscilloscope (to see ripple after the build), I have a linear transformer as power source and I do not want to oscillation ripple output. So that I want to conform from simulation.

I am going to make a bench power supply so that current is not fixed.
 
Here use my scope and try changing values.

You wont be able to use the 6A capacity even if the hFE was 100
The LDO resembles an ideal zener with <<1 Ohm=Rs so I used a 12V zener in this sim.

Raising the R1 reduces the power dissipation in all parts except the PNP as it raises only current 10% with a small Rb and much bigger I sense (R1)

Let's face it.

It's a bad design for high current. It isn't really a low dropout as the LDO suggests and is still very high ESR so load regulation error will occur.

As others have already reported, the Cap must be huge for high current using 50Hz. Try 50kHz (hint)

As I have stated in other questions, I choose RC1 = 9τ for τ=10ms or 100Hz with FW rectified 50Hz @ 10% ripple. ( it may be closer to 8τ, but close enough and give margin for aging)
C1 must also be low ESR type. $25
e36d201lpn472td79m_sml.jpg


With R (load)=5 Ω then C1 = 9*10ms/5Ω= 18mF

Bottom Line, Learn how to use an LDO as a Buck switching regulator (consult datasheet) better yet use a low Vth, very low RdsOn MOSFET bypass or better yet use a real Buck regulator chip. with good filtering.

The RdsOn of the bypass Pch MOSFET ought to be ~1% of the lowest load resistance at any desired low voltage.
 
Willen..........Regardless of doing a SIMULATION, you may still have oscillations in the Real Circuit.
A Simulation does not consider Ground Loops as can happen in Real Circuits.

Why not build up a LM723 Regulator circuit?
You can get both Voltage and Current Regulation, as well as a Good Low Noise Output.

Curious, What is the Voltage and Current Rating on your Transformer?
 
Last edited:
Willen..........Regardless of doing a SIMULATION, you may still have oscillations in the Real Circuit.
A Simulation does not consider Ground Loops as can happen in Real Circuits.

Why not build up a LM723 Regulator circuit?
You can get both Voltage and Current Regulation, as well as a Good Low Noise Output.

Curious, What is the Voltage and Current Rating on your Transformer?
Hi chemelec,
Thank you for the clarify. Sorry for the little abnormal part like LM723 or Buck converter as Tony said, I am living in such place where electronics is just trying to sprout. Till now I just saw TL431, 78XX and LM317 as regulator.

I am trying to utilize a transformer scrapped from old Black & White TV (used for 24V DC, 2A out?).
 
With 24 Volts @ 2 Amps Out from the transformer, you will Only get a regulated Supply Output of Just ONE AMP.
So there is no point to building that 6 amp regulator.

And an LM723 is a Common, but Older Part.
It should be easy to get one, Even in your country.
Probably a supplier in Kathmandu

Or just use the LM317 Without the added Transistor.
 
Last edited:
With 24 Volts @ 2 Amps Out from the transformer, you will Only get a regulated Supply Output of Just ONE AMP.
So there is no point to building that 6 amp regulator.

And an LM723 is a Common, but Older Part.
It should be easy to get one, Even in your country.
Probably a supplier in Kathmandu

Or just use the LM317 Without the added Transistor.
I crossed 27 years of age and TILL NOW I never been there in Kathmandu (our capital city). I studied in the eastern part of Nepal and completed university, though work never pushed me there to Kathmandu. Thousands of 'common parts' (for Canadian people) are 'rare parts' for Nepalese. As I know the small suppliers there in Kathmandu do not ship there product as home delivery. I just heard few shops of cloths or vegetables are starting to deliver their products to customers' house recent days, but just within the Kathmandu city. I am surviving 500km far east. :)

My intention to use the additional Q1 is to divide down the heat and power dissipation.
 
Q1 is NOT Required.
Even if you used Q1, it would require a Heat Sink on it.

So Just put a HEAT SINK on the LM317 and it will work good.

ALL Power Transistors and Most Voltage Regulators REQUIRE HEATSINKS to Help in Removing HEAT.
That is why the LM317 has a Hole to Bolt on a Heat Sink.
 
Can you salvage old PC's or TV's? then you have access to tons of electronics
 
To help understand LM317 , I made a simplified equivalent circuit. that you can change at will , (enable) JAVA in browser.
View attachment 95645

I did not add bypass current booster but that can easily be done using PNP or Pch FET so that current sense biases the bypass switch.

The 16V input is an ideal source with a switch add +/-2V of ripple or sag. I chose 0.1Hz but it could be 100Hz just to show effect of dropout when Vin-out<2V or so.

I chose fixed values of hFE, so they are shown as ideal transistors not real.

RED parts dissipate heat.

Any questions for the forum?

Tony
Hi
Now I watched your link and it's an amazing!!! cannot I save the page or something like that so i can use the simulation site while I am in offline mode? I have a computer but have no internet access. or any same type of free software? (to visualize every circuits)
 
There are two sources of engine execution, javascript and Java. It should be possible to use the source code for the Falstad site
permission of Paul Falstad the source project for this version of the application is now available on GitHub under a GPLv2 license.

https://github.com/sharpie7/circuitjs1

save that portion of the website to a flash drive then use it locally to run offline.

https://www.falstad.com/circuit/directions.html shows the basics of how to use it online

This programmer converted it to javascript instead of JAVA.
https://lushprojects.com/circuitjs/ This simulator makes extensive use of HTML5 features and definitely needs a modern browser. It also performs a lot of calculations in JavaScript and the speed of this varies a lot between browsers. Currently Chrome seems to have the best performance and feature support for this application. Internet Explorer also runs the JavaScript well but sadly lacks compatibility with all the file menu options.

It may not be the most efficient simulator or the most powerful, but it is based on physics but in the case of output drivers, caps and logic, they have zero ESR properties, so you must use experience and add a resistor and/or load LC parts to understand those effects for fast pulses.
Time scale can be changed but maximum frequency of clocks will be determined by the scope trace , execution interval in ns us ms etc. in the options menu.

Naturally a dual core minimum with least background junk running gives best performance on complex simulation so your browser will be running at max CPU. Some engineers on this site are challenged this way, but I know how to get more performance.
 
Willen if you want to make a decent linear PSU, scrap something with a hefty A.C. transformer. Perhaps an old microwave transformer or audio amp , even surround sound systems of about 10yrs vintage, rewind the HV secondary with a few turns of #12 to #16 AWG enamel wire to give u perhaps 18V or so RMS and you have the basis for a high current supply. This type of transformer is not very efficient at low loads but, hey, I've done it and it works fairly well. Do NOT use the transformer with the original HV secondary or there'll be no more Willen. Also discharge any caps on the HV side with an insulated pliers or screwdriver before yanking the transformer.

If you want to make it more efficient and cooler running, add about 100 turns to the 120VAC primary and then rewind the secondary for Low V. Also have a look at PTC current limiters to handle startup surges (can kill your bridge rectifier) which hi value, low ESR smoothing caps cause!
Before rewinding the secondary, wrap the transformer centre laminations very well with teflon 'plumbers tape' or kapton if u can get it. This prevents the sharp lamination edges from cutting thru the enamel insulation due to magnetic flux vibration in the laminations.
You can source enamel wire from another transformer , or anyone who repairs pumps, electric motors, fans, power tools etc. in your area. Or you can just 'pay' a motor rewinder to rewind the transformer for you, professionally. Just tell them the A.C. voltage you want out and to use the thickest gauge wire that gets you there.
 
Hi Mosaic,
It Sounds VERY efficient transformer of current capability. Currently I just have a 220V to 19V (AC) transformer, made for 24V Black and White TV. I need simple adjustable bench power supply so the transformer maybe suitable for now.
 
Ok, I guess you're running at 50Hz if it's a 220VAC xformer. Is it centre tapped on the secondary or primary?
What is your required output voltage range?
What is the max current required? As chemelec noted, you don't need hi current pass transistors IF your xformer can't supply the current.
 
I am a basic hobbyist for electronics so I experiment small circuits basically. I am wishing to get 20V max and around 1A or little more. (I tried to add a booster transistor so that if I got a bigger load later, so I do not need build another circuit again).


Xformer is simple two input (220V) and two output.
 
Fair enough.
I take it that you mean u have a single primary & secondary coil with 4 wires total on the xformer. Not two sets of two wire pairs.
The thing about a pass transistor is that it requires a base emitter dropping resistor to 'drive' it. This costs you current flow if the resistor isn't sized properly ( if the pass transistor isn't activated) and potentially causes MORE ripple due to voltage drop on the LM317 input.


Have a look at the 2nd 'hi current' option here:
http://www.reuk.co.uk/LM317-High-Current-Voltage-Regulator.htm

This takes a better approach, unloading the LM317 and loading as many pass transistors as you'd need. You can even add more later. If you're going for more than one pass transistor (for even higher current later on) it is imperative to keep them thermally linked on the same heatsink.
You'd still need 'smoothing' capacitors as shown in the first circuit on the page including the cap across the V-adjustment resistor. The bigger reservoir cap on the input. MikeMl/Tony Stewart showed how to calc. for capacitor sizing for ripple limiting. You'll have 100Hz ripple from a full wave bridge to deal with.

Here's a useful page on designing/using a 'choke' to make an (LC filter) along with capacitors for severe ripple management as occurs with higher current draw.
http://www.learnabout-electronics.org/PSU/psu12.php

http://www.aikenamps.com/index.php/chokes-explained

Automotive shops sell chokes for blocking engine ignition noise getting into the sound system. You'd need a bigger one than they offer OR place a few in series to get the required inductance.
 
adjustable-high-power-supply-with-lm317-voltage-regulator.jpg

This is circuit can self turn off if the battery full, trim the variable resistor until output voltage is 14,5V, the battery will almost full but safe
 
14.5 Volts is Acceptable in a Cyclic Charge. (Charge, Discharge, Charge again)
13.6 or 13.8 Volts is Acceptable in a Standby Charge. (Charging All the Time, While also using the battery)
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top