 Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me. |
Forum Index : Electronics : lm338 circuit voltage drop
Page 1 of 2   |
|||||
| Author | Message | ||||
| irishron40 Senior Member  Joined: 22/09/2014 Location: IrelandPosts: 251 |
Hi everyone, I am baffeled by the voltage drop under load. Load is 2.3A and @ startup it draws 2.7A. my transformer is able for 10 amp Rectifier is rated 25 Amp with no load it reads 14.8V with load voltage drops till 4 volt before ic is getting warm the lm338 is fitted on large heatsink I have tried different lm338 and caps thank you  |
||||
Madness Guru  Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
Might help if you posted an image that is readable. There are only 10 types of people in the world: those who understand binary, and those who don't. |
||||
| irishron40 Senior Member Joined: 22/09/2014 Location: IrelandPosts: 251 |
whats wrong with this drawing |
||||
| Pete Locke Senior Member Joined: 26/06/2013 Location: New ZealandPosts: 181 |
What voltage are you trying to get, and what is your input voltage? The input to these types of regulators need to be a MINIMUM of 3volts above the output. Otherwise the regulator is in saturation and won't control. Also, as a rule of thumb, the output capacitors should be 10uF per amp drawn. Cheers Pete'. |
||||
| irishron40 Senior Member Joined: 22/09/2014 Location: IrelandPosts: 251 |
thx Pete. I am building a hotwire foam cutter. The wire I use is ni-chrome. MY INPUT VOLTAGE IS ABOUT 17VOLT DIODES ARE RATED 3A THINK THEY ARE 1N5401 i LIKE TO REGULATE MY OUTPUT FROM 5-12 VOLT I JUST CANT SEE WHAT I DID WRONG WITH THIS CIRCUIT UNLESS LIKE YOU SAID IT COLD BE CAP SIZES. RON |
||||
| Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
It's literally the size of a postage stamp. even zooming right in it is barely readable. There are only 10 types of people in the world: those who understand binary, and those who don't. |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
The drawing is rather small... Anyhow, about a year ago I purchased a big bunch of LM338's from e-bay to use in some bench power supplies I wished to build at the time. Cannot now remember if it was 10 or 20 that I ordered, but they were all cheap Chinese. None of the damned things worked, and like yourself I was going mad with frustration. The circuit worked fine with some LM317's that I already had, but none of the LM338's would work in the same circuit. Ended up tossing the lot into the bin. Have had the same experience with TL431 regulators, bought a bag of a hundred of those, none a single worked. Later bought a batch of EPROMs none of which would program. So don't be too very surprised if you too have been suckered by some grinning Chinaman. Cheers, Tony. |
||||
Grogster Admin Group  Joined: 31/12/2012 Location: New ZealandPosts: 9308 |
17v input, 14v output, 2.7A current. Regulator is trying to dissipate: (17-14) x 2.7 = 8.1 Watts of heat - that's a lot for a TO220 package. I can't remember what the 338 is rated for, but I seem to recall all those linear regulators are 5W devices. I may well be wrong. I will now have to dig up the PDF so I can find out.  Why do you have a 10k/12k(can't read it) between ADJ and GND? Why not just the pot? That is probably part of your problem. I don't normally have an additional resistor there... Most adjustable regulator circuits have the 220/240 Ohm from output to adj, and a resistor from ADJ to ground to set the voltage. A variable regulator usually uses around 5k pot between ADJ and GND to vary the voltage. You should also be using 100n caps in parallel with your electrolytic caps. The 100n's filter off any high-frequency noise that might upset the stability of the regulator. Not quite so critical on the output of the regulator, but I always put one on the input and output just to be safe. Google Images is your friend:  EDIT: [Quote=Warpspeed]Anyhow, about a year ago I purchased a big bunch of LM338's from e-bay to use in some bench power supplies I wished to build at the time. Cannot now remember if it was 10 or 20 that I ordered, but they were all cheap Chinese. None of the damned things worked, and like yourself I was going mad with frustration. The circuit worked fine with some LM317's that I already had, but none of the LM338's would work in the same circuit. Ended up tossing the lot into the bin.[/Quote] Yep, the exact same thing happened to me. I think mine were LM350's, but they actually came in a regulator kit thing, but if you pushed more then 1.5A or so through them, the regulator went open-circuit!  That should not happen with a device rated for 3A or so....  I have a thread here on the forums about that at the time. Smoke makes things work. When the smoke gets out, it stops! |
||||
Chopperp Guru Joined: 03/01/2018 Location: AustraliaPosts: 1057 |
Data Sheets Here It may be getting too hot as G mentioned. Probably don't really need the big C on the output. I assume you've checked the protection diodes (Orientation etc) Should work unless a dud ChopperP |
||||
| Chopperp Guru Joined: 03/01/2018 Location: AustraliaPosts: 1057 |
Did you mean 10,000uF per amp? This will give about 1V P-P ripple. I assume you mean straight after the rectifiers. If you meant after the regulator itself, then please disregard my ramblings. ChopperP |
||||
| Pete Locke Senior Member Joined: 26/06/2013 Location: New ZealandPosts: 181 |
.... No, 10uF/amp on the OUTPUT side of the regulator. Primary side is a whole other set of numbers. |
||||
| Chopperp Guru Joined: 03/01/2018 Location: AustraliaPosts: 1057 |
@Pete OK, I hadn't heard of that one before. ChopperP |
||||
| Grogster Admin Group Joined: 31/12/2012 Location: New ZealandPosts: 9308 |
Neither had I, if it makes you feel any better ChopperP. I do know that the output cap is much smaller then the input cap, but I did not know about the 10uF per-amp thing. I tend to always stick 100uF on the output of most of my linear regulators. 1uF or so on LDO regulators. Again, depends on the datasheet for the device in question. Smoke makes things work. When the smoke gets out, it stops! |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
Should not be needed. The voltage regulator all by itself should regulate the output voltage. Only reason to fit a fairly small capacitor across the output might be to reduce the broad band noise for something like a super critical low level audio application. These are supposed to have internal thermal shutdown so even without any heat sink at all, you cannot burn one up if its working properly. The LM338's I had would not even run with zero output load. They were just all non working duds. Cheers, Tony. |
||||
| irishron40 Senior Member Joined: 22/09/2014 Location: IrelandPosts: 251 |
I WAS WONDERING FOR JST SCH A SIMPLE CIRCUIT. MUST TRY A 317 FOR THE CRAIC i ONLY WANT THIS TO HEAT UP A NICHROME WIRE FOR CUTTING FOAM AND FOR BENDING PERSPEX. i WANNA REGLATE IT AS FOR CUTTING FOAM I NEED LESS HEAT THE FOR BENDING PERSPEX. I MADE CIRCUIT BIGGER. I WONDER DO i NEED ANY CAPS AT ALL. RON  |
||||
| Grogster Admin Group Joined: 31/12/2012 Location: New ZealandPosts: 9308 |
Yes, you do. You need at least SOME kind of caps on the input, and usually, some on the output. As you are only running a resistor as a load(your wire), then you can probably ignore the output cap as Warpspeed says. But you still need some input caps. In your application, I would tend to go for 2200uF and 100n on the input, and nothing on the output as your load is purely resistive. Smoke makes things work. When the smoke gets out, it stops! |
||||
| Chopperp Guru Joined: 03/01/2018 Location: AustraliaPosts: 1057 |
OK with it G. Data sheets do give different recommended values for the OP cap depending on the device. The filter cap value for a full wave bridge rectified output at 100Hz is roughly 10,000uF per amp for 1V P-P ripple*. irishron40 has 10,000uF on the input so with 2-3A, he should get 2-3 V P-P ripple at the input to the regulator. This should be OK. * 2 formulas for caps. Q = I * t (Charge = current * time) Q = C * V (Charge = capacitance * Volts) therefore C * V = I * t or C = (I * t) / V for 50Hz rectified, F = 100Hz thus t = 10mSec (10^-2 sec) So if we set V = 1 & I = 1 C = 1 * 10^-2 / 1 F which = 10^-2 * 10^6 uF C = 10,000uF Note that there are some assumptions made to this reasoning  ChopperP |
||||
Tinker Guru Joined: 07/11/2007 Location: AustraliaPosts: 1904 |
If you want to bend perspex you need quite a different device. Hot nichrome wire cuts (melts) through foam. You do *NOT* want to melt perspex to bend it. I have a home made perspex bender here. It uses 10mm OD pyrex glass tubes which have spiral wound 240V elements inside. I think they were 200w each, connected in series. The heat was controlled by a simple light dimmer. The tubes heat the top and bottom of the perspex at the same time. If you heat only one side it takes a lot longer, you get a not so nice radius bend and you are limited to about 3mm max perspex thickness. Klaus |
||||
| irishron40 Senior Member Joined: 22/09/2014 Location: IrelandPosts: 251 |
THX EVERYONE. PLEASURE TO BE BACK LOL I REMEMBER MY FIRST POST WHEN I JUST STARTED LEARNING ABOUT ELECTRONICS. STILL LOADS TO LEARN BUT A FAR BETTER NDERSTANDING ABOT IT ALL NOW. RON |
||||
| Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
 There are only 10 types of people in the world: those who understand binary, and those who don't. |
||||
| Page 1 of 2 |
|||||
 Print this page |
