Page 1 of 2

Buyer beware may be LM338...... advice Welcome !

PostPosted: Mon Mar 16, 2020 6:08 pm
by Harry Dalek
I ordered 100 LM338 5amp regulators pretty much a drop in replacement for the LM317 least i think if i am not mistaken ?
https://www.aliexpress.com/item/3221549 ... 4c4dRjjqQI
i am using a few of those the 317's to drive my motor on the PhonoVision recorder which i wanted to replace with one of those .
Well my order came the other day and testing about 20 of them so far not one works ! as far as i know the Lm338 should work as is ..i was thinking perhaps the Pinouts are different ?
Any case looking at this video got me worried i have been ripped off but i am using the flat version


youtu.be/AvF7HyYklwI

OK looking at my version with the +P on the left what does the +P mean ? as its in a different place on the same regulator but different brands
LM338T-5-Amp-Adjustable-Regulator.jpg_350x350.jpg
LM338T-5-Amp-Adjustable-Regulator.jpg_350x350.jpg (20.73 KiB) Viewed 11132 times


These are other on line below with different positions for the +P ...I did experiment to see if the pin outs were different to no working result ....pretty much hooking it up as a lm317 there's no adjustment pretty much input voltage is same as output
583531_1.jpg
583531_1.jpg (35.66 KiB) Viewed 11132 times

AD283-2.jpg
AD283-2.jpg (49.93 KiB) Viewed 11132 times

Re: Buyer beware may be LM338...... advice Welcome !

PostPosted: Mon Mar 16, 2020 8:54 pm
by Steve Anderson
Here's the LM338 datasheet, this should confirm (or otherwise) the pin-out connections...

They should be a drop-in replacement (electrically) for the LM317, I haven't checked the pin-out though.

Steve A.

If they still refuse to work, you've either got fakes (as per the video) and/or duds.

No idea what the 'P+' means...

It's a reminder of the old adage, "If it seems too good to be true, it probably is."..and at US$16.99 for 100 it seems too good to be true to me...

In the UK these devices are around US$3.00 each in 100-off quantities (Farnell/Element 14)...i.e. US300.00 for 100. Compare is the byword,

Re: Buyer beware may be LM338...... advice Welcome !

PostPosted: Mon Mar 16, 2020 10:39 pm
by Harry Dalek
Steve Anderson wrote:Here's the LM338 datasheet, this should confirm (or otherwise) the pin-out connections...

They should be a drop-in replacement (electrically) for the LM317, I haven't checked the pin-out though.


Ok Steve looking at the data sheet it seems i am hooking it up correct dead copy of the lm317 pinouts .

I would not say its as bad a that can version in the you tube video as there was nothing in those sneaky bastards made .. hooking mine up as a lm317 just did not change voltage reversing input output it started to warm up so i would say faulty more than a dead fake as something is inside it .
i have only tested 20 so far it would be nice if one works !

If they still refuse to work, you've either got fakes (as per the video) and/or duds.

No idea what the 'P+' means...


Yes could be a fake still branded a regulator but it might be something else i will look into it more do some tests on them /

yes the P+ got me stumped i was thinking either input or output but its on the adjust pin seeing others are stamped in different places makes it worse !

OH well i will take this up with the site i used to buy them on and get my money back ...or send me the real things

It's a reminder of the old adage, "If it seems too good to be true, it probably is."..and at US$16.99 for 100 it seems too good to be true to me...

In the UK these devices are around US$3.00 each in 100-off quantities (Farnell/Element 14)...i.e. US300.00 for 100. Compare is the byword,


Yes true but a dangerous game to play in China if caught out ?

Re: Buyer beware may be LM338...... advice Welcome !

PostPosted: Tue Mar 17, 2020 11:10 am
by Andrew Davie
According to http://www.ti.com/lit/ds/symlink/lm338.pdf the "P+" means "Product Enhancement Tested" - !!

Re: Buyer beware may be LM338...... advice Welcome !

PostPosted: Tue Mar 17, 2020 1:22 pm
by Steve Anderson
Thanks Andrew, I'll follow that link and find out what it actually means...sounds like a marketing ploy to me.

Steve A.

Re: Buyer beware may be LM338...... advice Welcome !

PostPosted: Tue Mar 17, 2020 1:33 pm
by Steve Anderson
Harry Dalek wrote:Yes true but a dangerous game to play in China if caught out ?


As so many products coming out of China are fakes/duds/sub-standard they cannot have the manpower, means, laws, or will to enforce something similar to the UK's consumer protection laws. They probably wouldn't apply to exports out of China anyway.

Steve A.

Re: Buyer beware may be LM338...... advice Welcome !

PostPosted: Tue Mar 17, 2020 6:26 pm
by Harry Dalek
Thanks Andrew and Steve for your interest here ...Now i know Andrew thanks to your keen eyes ..i don't like the position of the +P very confusing ..more confusing it should of been tested before being sold bit of a fib here .
I grabbed a LM317 today just looked at it with my meter and compared it to the LM338's i randomly picked to test yesterday and showing faulty in the regulator circuit ..... with the meter some of those did show very low ohms between the adjust and output pins .

I now looked at some untouched LM338T's again by random and they showed they were very similar to the Lm317 i would think they should work ...Now either are going faulty in the circuit or by luck if you can call it that i picked the bad ones to test .

My bet is something is killing them inside them when powered up when i have some more time perhaps tomorrow i will one by one see if a good one gos in and a bad one comes out after powering up , well the LM317 works fine in the regulator circuit so it should handle it .

I am crossing my fingers i have some that are ok and i picked the bad ones yesterday but i know this seems unlikely ...Testing is the only way to find out .

I did contact the seller and the reply was i think what you can't use them ! i did explain perhaps google translate to Chinese is not as good when you tell them they are faulty !

Re: Buyer beware may be LM338...... advice Welcome !

PostPosted: Tue Mar 17, 2020 7:03 pm
by Andrew Davie
Harry Dalek wrote:Thanks Andrew and Steve for your interest here ...Now i know Andrew thanks to your keen eyes ..i don't like the position of the +P very confusing ..more confusing it should of been tested before being sold bit of a fib here .


They said it was tested.
They didn't say it passed the test(s).
No fibbing involved!

Re: Buyer beware may be LM338...... advice Welcome !

PostPosted: Tue Mar 17, 2020 7:06 pm
by Klaas Robers
It was long ago... about 1970. The 7400 TTL circuits were just on the market. There was an advertisement in my ham-radio magazine that I could buy "7400-type" circuits "untested". I was a student and wanted to make an electronic digital clock. So I bought one pack of 10 IC's of the type 7490, a divide by 10 chip. They had no mark "7490", but at the bottom there was only printed "90". I tested them one by one .....

- In fact all circuits did something.
- About half of them divided correctly by 10,
- may be the "0-set" or the "9-set" was not working,
- but for a clock, 3 stages (the "ones") don't need that.
- The 7490 has a divide by 2, and a divide by 5 function,
- which you have to inter connect externally,
- so some of them only divided by 5,
- others only divided by 2.
- I found one that counted: 0 1 2 3 4 5 6 7 0 1 2 3 4 5.....,
- but the reset worked,
- for a clock I needed 2 stages that have to count just to 6 and then reset.
- Bingo!
- One of the circuits counted 1 2 3 1 2 3 1 2 3....
- but could be reset to 0,
- and then counted 0 1 2 3 1 2 3.....
- For the tens of the hours this is just far enough.
- One of the circiuts had its "indication notch" at the bottom of the package,
- after bending all pins "upside down", it worked Ok.

So they were tested and failed. But they did something!! At the end I have used all circuits, built the clock and it is still working.!

Re: Buyer beware may be LM338...... advice Welcome !

PostPosted: Tue Mar 17, 2020 7:23 pm
by Andrew Davie
Many years ago I sold some old videogames on eBay. I couldn't be bothered testing so I sold them "as is, untested"
And the buyer was astounded they worked. He was sure that "untested" was a sneaky way of saying "tested, not working".

Re: Buyer beware may be LM338...... advice Welcome !

PostPosted: Tue Mar 17, 2020 8:37 pm
by Steve Anderson
Again, another adage, "You get what you pay for." I use a reputable semiconductor source here in Bangkok, many other sources are generally flooded with fakes/copies - call them what you like. I'd rather buy 10 that are certain to work than 100 that "might.' That applies from a simple BC547 up to a 32-bit microcontroller. Remember, I'm only an hour's flight from China, cheap export costs compared to elsewhere.

Would you get on board a Chinese made aircraft? Apply the same logic and decision-making.

Our neighbours bought a brand-new Chinese made SUV, within six months it was replaced with a second-hand Toyota. (I run a 30 year old Nissan diesel, over 400,000km, still safe and reliable).

Steve A.

Re: Buyer beware may be LM338...... advice Welcome !

PostPosted: Tue Mar 17, 2020 8:52 pm
by Harry Dalek

Re: Buyer beware may be LM338...... advice Welcome !

PostPosted: Tue Mar 17, 2020 9:06 pm
by Steve Anderson
Whoo, wait a moment there! The LM337 has no heatsink! It will go into thermal shutdown when the chip temperature gets near 125 Celcius. The regulator appears to be working OK at first. You do the math, with no heatsink the thermal shutdown will cut in. This is an example of 'Fake News."...and someone not knowing what they are doing...

I wonder how embarrassed he was when he bought a 'genuine' item and got the same result?

This is a case of RTFM (Read The F****** Manual), or in this case, the datasheet.

Steve A.

Re: Buyer beware may be LM338...... advice Welcome !

PostPosted: Tue Mar 17, 2020 11:45 pm
by Harry Dalek
Steve Anderson wrote:Whoo, wait a moment there! The LM337 has no heatsink! It will go into thermal shutdown when the chip temperature gets near 125 Celcius. The regulator appears to be working OK at first. You do the math, with no heatsink the thermal shutdown will cut in. This is an example of 'Fake News."...and someone not knowing what they are doing...

I wonder how embarrassed he was when he bought a 'genuine' item and got the same result?

This is a case of RTFM (Read The F****** Manual), or in this case, the datasheet.

Steve A.


Mine was on a heat sink but no indication of ever working no heating up at all no need to dissipate the heat output same as input voltage ...hes might of just been genuine then as you think .

Thats an interesting Story Klaas make use of what the ics can do after you knew what they were ,perhaps i should look into also powering my regulators low voltage just to see what the range of them are if any .

On your thoughts Steve on where and where not to buy i forgot about your dislike for Chinese parts i will from now on keep that in mind ..I forget there are people in the electronics parts world that will do this .

Re: Buyer beware may be LM338...... advice Welcome !

PostPosted: Wed Mar 18, 2020 2:11 pm
by Steve Anderson
Harry Dalek wrote:...i forgot about your dislike for Chinese parts...

Not quite correct Harry, many parts are made in China that are just fine. Those under supervision of a reputable company, Fairchild, Microchip and many, many, others with a proper Quality Control program. Those in the video are an obvious out-and-out scam (The TO3 packaged ones). The guy with the TO220 devices I think it is simply not having a heatsink - also no small (100n) caps on the input or output quite possibly leading to instability....especially with an inductive load (motors).

Those LM338's you bought, you cannot see if there's actually a chip inside without an X-Ray machine. Or simply they're all just duds.

Caveat emptor - Buyer beware.

Steve A.