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.

Lm388 1.3 - 30v power supply

Status
Not open for further replies.
Cars have been drive-by-wire for years. Isn't the gas pedal a potentiometer that could develop crackly bursts of full power from the engine?
 
???? Exactly why any good designer puts a lot of low pass filtering after the pot wiper and on anything that is a slow-changing reference voltage!

In the PSU example above, any noise or disconnect fault of the pot wiper (even a couple uS) causes the voltage difference between Vout and Vref to instantly drop to 0v. Do the math, the 5amp regulator chip immediately goes very low impedance and connects a massive cap charged to 30v directly to the load.

PSU reg chips are designed to have very fast transient response AND at full current. You would never find any pro adjustable PSU with no cap on the adjust voltage... Ever. Hence I said it was a critical part. :)
 
Can anyone get a current adjuster for lm338 or is there any shorciruit protection which is happening at its output due to various reasons, i have lost almost 5nos of lm338
 
Can anyone get a current adjuster for lm338 or is there any shorciruit protection which is happening at its output due to various reasons, i have lost almost 5nos of lm338
An LM338 made by National Semiconductor who invented it works perfectly.
Maybe you have a cheap Chinese copy?
Maybe you are driving an inductive load but forgot to install the protection diode that clamps the inductive voltage spike?

The LM338 by National Semi limits the current above 5A. It shuts down if it gets too hot.
 
The current limit of the LM338 is internally limited as Audioguru stated. The minimum according to the data sheet is 5A. If you want to make it less than that it can be done with some external components. Usually a low value resistor and a transistor like a 2N2222.
 
You are probably killing it with spikes. They are hard to kill with over current or over heat as both cause a temperoary safe shutdown.

But overvoltage on Vin, and/or high energy spikes on the Vout pin will kill it quickly. What is your Vin voltage? And what is the load on the output of the PSU (and how are you connecting/disconnecting that load?
 
I am using only National Semiconductor LM338K and using it as desktop powersupply, (0-12 volt, 9 amps transformer) vin is 18 volt after bridge rectifier & capacitor and using it for checking all the circuit board which ever am servicing. Seperate toggle switch is provided for + output for connecting and disconnecting the load. Protection diode is also fixed at output terminal.

Majority of time i had noticed is when the + probe and - probe (output of the powersupply) accidently touches eachother the out put is "0" volt on touching the ic the temperature is also normal (cool). at some circumstance it may be difficult to find the amper of the item which is under service, so i'll have to connect to the powersupply if it is with in 5 amps no problem if not........ the output is only "0" volt.

As Mr.RB says if it is temperoary safe shutdown what will be the approxmiate duration to get back to normal

Thanks
 
If you short the output of your LM338 power supply then the LM338 will try to pass 9A which will drop to 6A in 1/10th of a second. With 18V across it then it dissipates 108W to 162W and will shut-down when it gets too hot internally.
It might take 30 seconds or more for it to cool enough to begin working again.

It can dissipate about 60W continuously with a huge heatsink.
 
The heat sink is so heavy what i have provided for lm338. and hours later even i checked after the short occured still no output.

I have designed a short circuit protection for lm338 to fix this problem, can i incorporate this ? any suggetion or comment pls.

Thanks
 

Attachments

  • LM338 POWER SUPPLY.JPG
    LM338 POWER SUPPLY.JPG
    137.8 KB · Views: 1,430
That seems very complex. :(

What are you using the power supply for? A 12v battery charger? Driving big 12v motors?

I would think about adding something to the output terminals to stop any spikes, like maybe a 20v 5 watt zener.

Could you please show your schematic of the PSU as it is now, and explain exactly what the 12v 5+ amp loads are that yopu connect to it.
 
I'm gonna guess slamming the thermal shutdown of the LM338 is not good for the longevity of the chip.
 
I know the circuit is little complex but still will it serve the purpose

Clients servicing items may include motors, circuit boards motors linked with ciruit boards etc. Mostly all are Industrial related it can be of any operating voltage like 12V,9v,5V,6V, etc.

power supply i tryed both circuits which is enclosed


I am using only National Semiconductor LM338K and using it as desktop powersupply, (0-12 volt, 9 amps transformer) vin is 18 volt after bridge rectifier & capacitor and using it for checking all the circuit board which ever am servicing. Seperate toggle switch is provided for + output for connecting and disconnecting the load. Protection diode is also fixed at output terminal.

Majority of time i had noticed is when the + probe and - probe (output of the powersupply) accidently touches eachother the out put is "0" volt on touching the ic the temperature is also normal (cool). at some circumstance it may be difficult to find the amper of the item which is under service, so i'll have to connect to the powersupply if it is with in 5 amps no problem if not........ the output is only "0" volt.
 

Attachments

  • 5 amps.jpg
    5 amps.jpg
    51.6 KB · Views: 687
  • 10 amps.JPG
    10 amps.JPG
    39.7 KB · Views: 869
Ahh I thought you might be attaching it to motors... :)

I can see a couple of problems. Your Vin cap is way too big, in the event of a Vout short a 10000 uF cap will allow massive surge current through the LM338 before it goes into limiting. Also you have no Vout cap, this is a real issue when driving things like DC motors that produce spikes from the commutator and/or motor inductance.

I would reduce the Vin cap to 2200uF or so, the LM338 will have good input ripple rejection so you really don't need so much capacitance on Vin. It won't hurt to put a resistor between the first cap and Vin, this will reduce possible current surges. Maybe drop 0.5v at 5A full load so that needs about 0.1 ohms 5W or 10W.

I would add a minimum of 2200uF to the output. This goes a long way to absorbing any transients on Vout caused by your loads instead of destroying your LM338.

My 5.5 amp bench supply is set up with the caps like this. It's not datasheet typical but it has worked without failure for 25 years and it's driven some real nasty motor and inductive loads and lots of short circuit and leads arcing etc etc. Don't be afraid to put a big Vout cap on a supply if it will get nasty loads like motors. You may want to put a load resistor across the output cap too, I think my supply has a 100 ohm resistor there.
 
door intercom

i'v little experience in electronics and now i'v been asked to design and build a loudspeaker door intercom. i'v tried searching 4 related circuit but found none. any one with idea on this shld pls help me by mailing me via abbey4evera@yahoo.com. thank u.
 
Thanks Mr RB , nessary corrections has been made in the new schematic pls verify for any errors
 

Attachments

  • 5 amps corrected.jpg
    5 amps corrected.jpg
    62.5 KB · Views: 523
Yep that should take a lot more punishment from your DC motors. :)
I would also suggest you make sure you have rubber boots on your alligator clips, to stop them touching each other and also it's a good idea to cut one lead an inch shorter than the other so when the leads flop the 2 clips won't touch together, that will reduce shorts too.

But there are people here with a lot more power supply knowlege than me, maybe someone will speak up with further ideas...

Since you are connecting "mystery loads" it might be worth putting another LM338 as a switchable current regulator before your main LM338 voltage regulator (if you have enough voltage headroom, current regulators normally take about 3 volts).
 
The 2200uF main filter capacitor is much too small for a current of 5A. It should be at least 10,000uF or the output will not be regulated because it will be full of ripple.
 
Status
Not open for further replies.

Latest threads

Back
Top