Defective/ fake LM317T

[Maarten]

Originally Posted by MrBungle

Both look fake to me. None of my ST ones (from RS) look like that. I tried to take a photo of some but failed miserably.

On the ST ones the numbering is a different font (more "robotic") and there's a G next to the (e3) bit at the bottom right. There is no exposed metal in the recesses on either side either.

The robotic font is newer. The right one could pass for a somewhat older stock example, but not 100% sure, I would have to look for more examples.

[MrBungle]

Originally Posted by Terry_VK5TM

R1 at 510 Ohms is more than double the recommended value of 240 Ohms and the pot only needs to be 5k to give a 2 - 30 v range.

True. The 240 ohms is a bit of a misnomer too though. Someone copied it from the datasheet decades ago, which was honestly pretty poorly written in context to how people tend to use the things, and it became religion.

A neat trick to making calculations easier is to use a 120 ohm resistor there. This does two things:

Firstly it makes the other resistor value easy to calculate in your head as the reference adj is 1.25 volts. So for 12 volts you just take 1.2 volts off and multiply by 1000 then round down to the nearest E12 so 1K. This pops out around 11.7 volts, slightly under intentionally, so you can then slide a 100 ohm trimmer on the end to give 11.7-12.7 volts for precise trimming or use a precision E24 in there instead.

Secondly this divider also sinks 10mA which is the minimum required DC load current for regulation. That means it'll be quite happy with whatever is or isn't connected to the output. With a 240 there it's only about 5mA which is borderline unstable with about a 10% chance of doing something weird.

There are some real cock ups with the LM317 out there. QRP labs sell a few ...

[ajgriff]

This is one of those threads that has generated so much noise that it's hard to see the wood for the trees. To my (simple) mind the following chronological extracts and the attached image crop may hold the clues needed to solve the mystery:

Originally Posted by Jolly 7

The kit did not come with the insulated washer or mica sheet either. Removing the IC has caused damage to the board and it looks like it could be a write off.

Originally Posted by Electronpusher0

It looks like the pot is just being used as a variable resistance, the bottom 2 pcb pads for the pot look joined together, in which case you only need 2 wires.

Peter

Originally Posted by Jolly 7

Replacing the 10K linear pot with a 22K linear one and turning it clockwise gave acceptable voltage readings till around midway, but bizarrely the readings started to fall after that to reach 1.2V again.

Originally Posted by ajgriff

It looks to me as if the 'flying pot' might have been wired with the track ends connected to the negative rail and the slider to the Adj pin.

Alan

If I were the OP I would:

1. Remove the flying pot with its leads and put back the original.

2. Reinstate the supplied LM317 assuming it too hasn't been fried.

3. Make sure that any circuit board damage is made good and that there are no shorts or discontinuities.

4. Try again.

Alan

[Jolly 7]

The aim of mounting the pot the way I did was to make it panel mountable and make the entire unit fit in a case. I would not have been able to do that with the supplied blue pot with a tiny screw for voltage adjustment. In relation to the wiring of the pot, the PCB already came with a tracing linking the wiper to one end of the track.
I selected a 22K panel mountable pot because I have no more 10K ones, at least for now. I also suspect my 10K pot may have been burnt by the IC, because it was reading in the megaohm range when I tested it after desoldering.

[Jolly 7]

@ajgriff : here is what I am hoping to achieve in the end. The picture is of a unit I built last year from a preassembled regulator module but it is too noisy for radio circuits, hence the new LM317T circuit

[G6Tanuki]

As others have noted, you need a certain minimum current through the resistor-network; I've generally used a 5K variable, *and* a separate load-resistor from the +ve output to ground to place a consistent load on the chip.

The diodes and decoupling caps are highly recommended too!

[I've used loads of the TO3-cased LM117/LM317 3A-rated "Steel" regulators in the past and they behave well if treated well; I've also built quite a few little power-supplies using a Vellemann kit

http://www.velleman.co.uk/contents/e...198_k1823.html

based on the TO220 LM317 and they work just fine for powering Roberts radios like the 707 and 606].

[ajgriff]

Originally Posted by Jolly 7

The aim of mounting the pot the way I did was to make it panel mountable and make the entire unit fit in a case. I would not have been able to do that with the supplied blue pot with a tiny screw for voltage adjustment. In relation to the wiring of the pot, the PCB already came with a tracing linking the wiper to one end of the track.
I selected a 22K panel mountable pot because I have no more 10K ones, at least for now. I also suspect my 10K pot may have been burnt by the IC, because it was reading in the megaohm range when I tested it after desoldering.

I fully appreciate why you ultimately wish to use a different pot rather than the original trimmer. However, I was trying to understand the odd behaviour described in your post #20 (third quote in my earlier post). It still appears to me that your flying pots were incorrectly wired and I can see no reason to change that view. I may be completely mistaken of course.

Alan

[Jolly 7]

Well, here's finally an LM317 that's not fake. I built a fully functioning voltage regulator from it and some other parts, including a scrap VCR transformer referred to in another thread, a heatsink from an old Woolworths wall wart and a special 220 ohm resistor (reading 250 ohms) from BVWTM ! It works fine with both a 6V radio circuit and a 24V motor.
Many thanks to everyone who helped me build this regulator properly from the scratch and without a kit !! The kit that failed earlier on has been repurposed into a fixed 6V regulator.
It works fine with a 'flying pot'. I selected a 5K one, a better wirewound Bourns one is on its way.