Author Topic: Increasing LM317 current  (Read 3036 times)

0 Members and 1 Guest are viewing this topic.

Offline clay1905Topic starter

  • Regular Contributor
  • *
  • Posts: 83
  • Country: au
Increasing LM317 current
« on: May 18, 2020, 07:54:30 am »
Presently I'm trying to construct a power supply for a portable valve radio. The original power transformers' secondary went open.
The schematic I've been following is here:

https://www.vintage-radio.com/projects/battery-set-psu.html

It's also below, but not mine and I can't give credit.

There's a catch, isn't there? Not quite enough current for the filaments. After about 5 to 7 minutes the radio fades away. At first I blamed some aging component in the radio, but when I burnt my finger on the LM317, I realised almost, but not quite enough current was being passed by the regulator, and it was gradually warming up until the thermal protection cuts in.
So, for a learner, a massive brain fart occurred. My schematic for increasing current capacity is included below. It's simple really. Just use two regulators in a parallel set-up with rather small resistances to ensure they stay pretty well balanced.
So why wouldn't this work? Nothing is ever this easy, and I'm not seeing anything like this idea on the web, so where has my thinking gone wrong?
Your help in understanding would be appreciated.

I've labeled the little balancing resistances as being 0.5 Ohms each, but that's just a nominal value. The schematic isn't intended to contain all the precise details just yet.

Thanks,
Clay.
« Last Edit: May 18, 2020, 08:06:08 am by clay1905 »
It's OK. We've all smelt like that since it happened.
 

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 9410
  • Country: gb
Re: Increasing LM317 current
« Reply #1 on: May 18, 2020, 09:08:40 am »
What sort of heatsink do yo have on the LM317?
Best Regards, Chris
 

Offline rdl

  • Super Contributor
  • ***
  • Posts: 3665
  • Country: us
Re: Increasing LM317 current
« Reply #2 on: May 18, 2020, 09:19:13 am »
Have you measured the actual current your circuit is supplying? Do you have a heatsink on the LM317?

Paralleling 3 terminal regulators is not a good idea. Two possible options, switch to LM350 or LM338 if you just need a little more current or if you need a lot more use a PNP transistor to boost current (look in a datasheet for the circuit).
 

Online Ian.M

  • Super Contributor
  • ***
  • Posts: 12806
Re: Increasing LM317 current
« Reply #3 on: May 18, 2020, 11:57:12 am »
Your resistor divider is chosen for 7.5V DC out, yet you say you want 1.4V DC out . . . .

Also, you cant add current sharing resistors to LM317 as regulators like that as they will cause the output voltage of your circuit to drop by 1.5V per amp of load current!

Furthermore the input voltage is  grossly excessive so its no wonder the regulator is overheating and going into thermal limiting.

IMHO it needs a complete  redesign,  e.g. take a toroidal transformer with a single secondary, then add an overwind to get a low voltage secondary for a much lower voltage at the regulator input + use a proper LDO so you don't need as much voltage drop across it, and also choose one rated for the required load current.   Set it up with a resistor dummy load for the max. current required, then adjust the overwind turns count for just enough input voltage to the LDO to eliminate ripple break-through, when running from a Variac (or fixed buck autotransformer) at 90% of  your line voltage, so its got enough margin for brownouts when fed direct from the mains.   N.B. the LDO is still going to need a decent sized heatsink . . .
« Last Edit: May 18, 2020, 12:00:25 pm by Ian.M »
 

Offline 2N3055

  • Super Contributor
  • ***
  • Posts: 6446
  • Country: hr
Re: Increasing LM317 current
« Reply #4 on: May 18, 2020, 01:05:11 pm »
Your resistor divider is chosen for 7.5V DC out, yet you say you want 1.4V DC out . . . .

Also, you cant add current sharing resistors to LM317 as regulators like that as they will cause the output voltage of your circuit to drop by 1.5V per amp of load current!

Furthermore the input voltage is  grossly excessive so its no wonder the regulator is overheating and going into thermal limiting.

IMHO it needs a complete  redesign,  e.g. take a toroidal transformer with a single secondary, then add an overwind to get a low voltage secondary for a much lower voltage at the regulator input + use a proper LDO so you don't need as much voltage drop across it, and also choose one rated for the required load current.   Set it up with a resistor dummy load for the max. current required, then adjust the overwind turns count for just enough input voltage to the LDO to eliminate ripple break-through, when running from a Variac (or fixed buck autotransformer) at 90% of  your line voltage, so its got enough margin for brownouts when fed direct from the mains.   N.B. the LDO is still going to need a decent sized heatsink . . .

His divider is configured for 1.4 V.
 
The following users thanked this post: WattsThat

Offline coromonadalix

  • Super Contributor
  • ***
  • Posts: 5795
  • Country: ca
Re: Increasing LM317 current
« Reply #5 on: May 18, 2020, 01:22:01 pm »
 
The following users thanked this post: Davidwasalreadytaken

Online Ian.M

  • Super Contributor
  • ***
  • Posts: 12806
Re: Increasing LM317 current
« Reply #6 on: May 18, 2020, 01:49:03 pm »
His divider is configured for 1.4 V.
No it isn't.

At zero load current, the LM317 regulators develop 1.25V (nominal) across the upper (100R) resistor in the divider, for a current through it of 12.5mA.  Neglecting the Adj pins currents, the same current flows through the lower (500R) resistor.  The total resistance is 600R, so from Ohm's law the  output voltage is therefore 600R * 12.5mA = 7.5V
« Last Edit: May 18, 2020, 01:51:54 pm by Ian.M »
 

Offline 2N3055

  • Super Contributor
  • ***
  • Posts: 6446
  • Country: hr
Re: Increasing LM317 current
« Reply #7 on: May 18, 2020, 02:19:07 pm »
His divider is configured for 1.4 V.
No it isn't.

First schematic.. 220 Ohm/27 Ohm
 

Online Zero999

  • Super Contributor
  • ***
  • Posts: 19344
  • Country: gb
  • 0999
Re: Increasing LM317 current
« Reply #8 on: May 18, 2020, 03:03:22 pm »
His divider is configured for 1.4 V.
No it isn't.

First schematic.. 220 Ohm/27 Ohm
The second schematic is larger and more eyecatching so it's no surprise it got more attention. It's also wrong: the LM317 will smoke if directly supplied with 15VAC.

The correct way to boost the output current of the LM317 is with a PNP pass regulator, as linked to prviously. Unfortunatly the overcurrent protection of the LM317 is lost, so extra components are required to make it short circuit proof.
 

Offline rdl

  • Super Contributor
  • ***
  • Posts: 3665
  • Country: us
Re: Increasing LM317 current
« Reply #9 on: May 18, 2020, 03:13:36 pm »
I'm betting either a heat sink or an LM350 (or both) is all that's needed.
 
The following users thanked this post: Davidwasalreadytaken

Offline schmitt trigger

  • Super Contributor
  • ***
  • Posts: 2205
  • Country: mx
Re: Increasing LM317 current
« Reply #10 on: May 18, 2020, 03:40:52 pm »
Use a switchmode supply.

As others have mentioned, your raw DC voltage is way too high for a low voltage, high current output. Any linear regulator you employ will dissipate several times more heat than the vacuum tube's heaters themselves!
 

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 9410
  • Country: gb
Re: Increasing LM317 current
« Reply #11 on: May 18, 2020, 05:55:17 pm »
I'm betting either a heat sink or an LM350 (or both) is all that's needed.

Agreed. I get the impression that the OP hasn't got any sort of heatsink fitted.

An LM317 should be easily able to handle the worst case dissipation without any current boosting. The fact that it ran for a few minutes indicates that it isn't pulling anywhere near 250mA anyway. Those little directly heated 1.5V valves in portables were very frugal.
« Last Edit: May 18, 2020, 05:58:17 pm by Gyro »
Best Regards, Chris
 

Offline drussell

  • Super Contributor
  • ***
  • Posts: 1855
  • Country: ca
  • Hardcore Geek
Re: Increasing LM317 current
« Reply #12 on: May 18, 2020, 06:54:24 pm »
Trying to run an LM317 at about 1.5v out from a ~20V DC supply is going to be an issue, even with trying to add some series resistance in front of it.

That 15VAC winding feeding the filament side is way too high. 

In order to continue to use that transformer you're going to need to increase the value of the series input resistor significantly and use a huge heat sink, drop the input voltage with a long string of diodes, or otherwise pre-regulate / drop some serious voltage before the LM317 so it doesn't have to drop all the volts by dissipating the extra power all by itself.  :)
 

Offline drussell

  • Super Contributor
  • ***
  • Posts: 1855
  • Country: ca
  • Hardcore Geek
Re: Increasing LM317 current
« Reply #13 on: May 18, 2020, 07:02:57 pm »
Agreed. I get the impression that the OP hasn't got any sort of heatsink fitted.

An LM317 should be easily able to handle the worst case dissipation without any current boosting. The fact that it ran for a few minutes indicates that it isn't pulling anywhere near 250mA anyway. Those little directly heated 1.5V valves in portables were very frugal.

Hmm, yeah I hadn't really thought about the fact that those battery tubes have very low current filaments, I had in my head the more typical filament requirements for, say, 6.3v tubes.   Even at 250ma the LM317 should only be dissipating about 5w even without the series resistor, (right?) so then that should be doable with a moderate heatsink.

The OP needs to let us know what their current heat sink is, I suppose.  (if any)  :)
 

Offline edpalmer42

  • Super Contributor
  • ***
  • Posts: 2260
  • Country: ca
Re: Increasing LM317 current
« Reply #14 on: May 18, 2020, 07:06:48 pm »
Another way to deal with the overheating would be to increase the value of R4.

The output of the rectifiers is going to be ~20V peak.  The existing R4 (22 ohms) will drop ~5V5 @ 0.25A which results in an input voltage to the LM317 of ~15V.  So the LM317 has to drop that to 1V4 and dissipate 3.4W.  If you increase R4 from 22 to 68 ohms, the LM317 will see an input voltage of ~6V and will only have to dissipate ~1.2W.  R4 would have to be rated for at least 5W.  If the radio draws more or less current than 250 ma, adjust the new value of R4 to give the LM317 an input voltage of 2 - 3 volts more than the output.

This might allow you to either retain the existing heatsink, or add a smaller one that is easier to add to the circuit.
 

Offline clay1905Topic starter

  • Regular Contributor
  • *
  • Posts: 83
  • Country: au
Re: Increasing LM317 current
« Reply #15 on: May 18, 2020, 11:39:52 pm »
Well, thanks for all the input.

Firstly, there's a typo made in haste. The input is wrong. With the resistors mentioned below, the voltage on the input leg of the regulator is 4.3VDC. This *should* be just adequate headroom.

R4 in the full schematic; There is a bit of discussion on the web site I pointed to about it being inadequate to dissipate enough power.  The 22 ohm resistor was found lacking by a previous builder, and I have 2 of 27 Ohms & 5 Watts in series. This drops the voltage to 4.3VDC without too much dissipation per resistor. I would have used 2 Watt jobs, but they're hard to source here.

The divider network in the full schematic is not what I'm using. I will be using the values in the little sketch, of 100 and approx 500 Ohms.

the heat sink is a"U" shaped piece of finned aluminium that was bought for the purpose, but probably insufficient. The back of the U shape is about the same area as the back of the chip, and the two sides are also about the same size, less the slots. There 's plenty of room in the cabinet, so a bigger heat sink wouldn't be an obstacle.

Current. No, I haven't measured, but that'll happen after breakfast. It should not be much more than 350 mA when stabilised as there are effectively 6 filaments in parallel, each drawing ~50mA.
The datasheet says the LM317 should be capable more than this. A larger sink may be all I need from here.
Went off half cocked here, I'm afraid. When I measure the actual current drawn I'll be in a better place to assess whats next.

Thanks again everybody,
Clay.
« Last Edit: May 19, 2020, 12:06:27 am by clay1905 »
It's OK. We've all smelt like that since it happened.
 

Offline rdl

  • Super Contributor
  • ***
  • Posts: 3665
  • Country: us
Re: Increasing LM317 current
« Reply #16 on: May 19, 2020, 03:21:27 am »
Sounds like you are almost there. Don't overlook thermal coupling of some type between the regulator and the heat sink, it makes a big difference. The silicone pad type is okay, not as good as actual heat sink grease but much less messy.
 

Offline clay1905Topic starter

  • Regular Contributor
  • *
  • Posts: 83
  • Country: au
Re: Increasing LM317 current
« Reply #17 on: May 19, 2020, 07:59:15 am »
Indeed. I have a monstrously over sized heat sink I snaffled from a defunct T.V. set. Yes, I'll be putting a smear of that evil thermal paste between the regulator and the heat sink, as well as getting it on the upholstery, the curtains and probably inside the 'fridge too.
I'm not worrying about a mica insulator as the heat sink won't be in a position to cause trouble with a short.

In any case, I think I'm well on the way to having a constant power supply, so I can get on with the alignment of the radio chassis soon.

Clay.
It's OK. We've all smelt like that since it happened.
 

Offline schmitt trigger

  • Super Contributor
  • ***
  • Posts: 2205
  • Country: mx
Re: Increasing LM317 current
« Reply #18 on: May 19, 2020, 06:31:50 pm »
Last but not least;
How far away are the tubes from the regulator?

They may significantly heat up the surrounding ambient.
 

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 9410
  • Country: gb
Re: Increasing LM317 current
« Reply #19 on: May 19, 2020, 06:51:21 pm »
One think that strikes me as a bit dodgy... Originally, the radio would have been powered by two separate batteries. The power supply introduces a current path between the two supplies. Maybe it varies from radio to radio, but I would have though that this might cause some issues with the directly heated filament cathode biasing (unless both batteries were always conmmoned at the chassis).

Ideally you would use a transformer with two secondaries to maintain the original 'separation', maybe benefiting from more appropriate voltages too.
Best Regards, Chris
 

Offline drussell

  • Super Contributor
  • ***
  • Posts: 1855
  • Country: ca
  • Hardcore Geek
Re: Increasing LM317 current
« Reply #20 on: May 19, 2020, 08:16:54 pm »
One think that strikes me as a bit dodgy... Originally, the radio would have been powered by two separate batteries. The power supply introduces a current path between the two supplies.

As drawn, the original schematic has them separate, running from two different secondaries on the transformer.
 

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 9410
  • Country: gb
Re: Increasing LM317 current
« Reply #21 on: May 19, 2020, 08:24:42 pm »
Ah, sorry, you're right. I was thinking the high secondary voltage was being imposed by a shared secondary.

This is one of those times where the 'salvage and hoard' habit comes into its own - building a collection of multiple voltage secondary transformers!
Best Regards, Chris
 

Offline clay1905Topic starter

  • Regular Contributor
  • *
  • Posts: 83
  • Country: au
Re: Increasing LM317 current
« Reply #22 on: May 20, 2020, 08:26:53 am »
The transformer used supplies two independent 15VAC outputs. For many designs, this is essential as I understand things. For "transformerless" designs, I don't think it makes much never mind.
The chassis I'm presently looking at has the usual construction, variable capacitor, big inductors, switches and valves above the sheet metal chassis base. Other discrete components below the sheet metal chassis, and then the added refinement of a shield to enclose the lower section. The power supply is mostly going to be mounted on the underside of the shield closure. The transformer fits quite well where the original one stood. It all makes for a rather neat looking modification.

It's an interesting idea, the regulator getting heated by the filaments, but no on two counts in this instance. As described, there's a bit of vertical distance between the valves above and the regulator below, and these 1.4 volt filaments hardly get warm. Even after some time the glass envelopes are only just warm to the touch.

Clay.
It's OK. We've all smelt like that since it happened.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf