Rebranded IC? 14 pin LM13700, anyone have a pinout ? :)

[Ronan]

This is not a photoshop job, nor a prank. One of an order of 10 x LM13700's has the incorrect number of pins...



Hmmm  

[Scruffie]

  wow, laziest fake chipping ever.

They all share the same code stamp so i'd assume the others don't work.

[oldschoolanalog]

Please let us know who/where you got those from. That might save headaches for others at some point.
Thanks. 

[Ronan]

The seller was leemoom611, who has done around 20,000 sales with 99.2% positive feedback.
The sale date was exactly one year ago, I don't know if I've tried any of these IC's or not. I do know I breadboarded a LM13700 wah pedal type circuit, but going through my breadboards I cannot find that LM13700 circuit, or any sole LM13700 anywhere, I may have put it back with the other 10 I bought, but I can't be sure. I guess I'm going to have to test one to see how it goes. As for the 14 pin one, who knows what's inside that.

[oldschoolanalog]

http://www.ti.com/lit/ds/symlink/lm13700.pdf
It is (well, it should be...) a 16 pin IC. I looked around and haven't seen a 14 pin version. 
If you're not going to return them clean the top of one with some acetone. Maybe the original markings are under the blacktopping.

[Toney]

Awwww, they look exactly like the "National" LM13600 and LM136700's that are sitting in my parts box.
Got them from Polida last year before I wised up to his fake Jfets. Just kinda "hoping" those were OK. Haven't tested them yet.
I wonder what they could fake those with and get passable functionality? (in cases where all the pins are present )

[oldschoolanalog]

 >I wonder what they could fake those with and get passable functionality? (in cases where all the pins are present ...)<
They probably couldn't care less if they don't work. They have the customers money and most folks wont bother returning them due to the shipping cost.
Please consider cleaning the top of one to see what's really under there.

[Ronan]

I tried to clean the tops with nail polish remover, but it really looks like the tops of the chips have been shaved, and one of them has a tiny ridge along the side where the "shaving" didn't extend across the entire width of the chip. Then they appear to have been laser etched with very good detail/accuracy.

So does it work? Yes, it does. I breadboarded the fig 4 circuit in the posted datasheet, Vin/Vout = 940 x Iabc and outputs matched within 0.3dB (I take that to be around 3 or 4% output voltage match). I used 2 x 9V batts, RG's quick and dirty oscillator and a fluke scopemeter, so pretty rough and ready but it looks like the chip performs within spec. It does what the datasheet says it should do. I only tested one of the IC's, and put a 2nd one in for a quick check, it seemed OK.

Which takes us back to what Toney (Rod) said, "I wonder what they could fake those with and get passable functionality?" This is quite a mystery.

[armdnrdy]

Greedy people turn to desperate measures to make money regardless of how it affects the person down the line.

That is why I have been driven to design a BBD tester/authenticator.

If what you say about this IC performing within spec is indeed correct, I would test some aspects of its design of it a bit further.

The only difference between the LM13700 and the LM13600 is the bias in the buffer section. Check the data sheets for particulars. I have a feeling that if these ICs were re-labeled, but they appear to work within spec, they are relabeled LM13600s.

You might ask, why would they go through the trouble of relabeling an IC of this low value?

I don't have an answer accept for, why do they re-label any IC?

Maybe they found a huge lot of LM13600s but saw a market for LM13700s.

Check the buffer section.

[Ronan]

Unfortunately I don't understand the bias arrangement in the LM13600 which affects the buffer stage. I can see the difference in the internal schematics on the datasheets, but don't know a way to test it for its presence. If someone can suggest a way to test it to determine if the chip is a LM13600 vs LM13700, then I would be more than happy to carry out the test (or tests).

[R.G.]

So does it work? Yes, it does. I breadboarded the fig 4 circuit in the posted datasheet,

What pins did you hook to where? You only had 14 pins to hook to 16 places. Which two were left out?

The LM13600 and LM13700 have identical 16 pin layouts.

[Ronan]

Sorry, I didn't make it clear. I definitely did not test the 14-pin IC, I have no idea what is inside that, that one is a write-off. I just tested the other ones with 16 pins (2 off).

I discussed it with my wife, she suggested the Chinese may have bought a heap of surplus from some well-known manufacturer (e.g. Sony, Panasonic) who had their own brand name on the chips, and one of the conditions of resale was to relabel them. A possibility?

[R.G.]

OK - that makes sense.

It is possible that some company found some private-labeled LM13x00 chips and relabeled them, just as she insightfully picket up. It's probably more like they found out what the chips really were and did the relabelling on their own, or found unused stock from a company doing manufacture for someone else who knew what they were.

But it's impossible to tell. If it looks like a duck, walks like a duck, and quacks like a duck, it can be assumed to be a duck for engineering purposes.

At least until it grows horns. 

[Ronan]

OK, I'll report back if I see any non-duck behaviour

[Toney]

 OK assuming it is a sanded and branded house numbered LM13x00 of some kind, and you say its functioning as expected, do you have access to a "real" LM13600 or 700 to A/B it with?
Other than setting it up on a breadboard with a known chip and checking pin by pin, I don't know any way of testing these. ... well other than the obvious go/no-go in a circuit. I am keen to hear any results you have as I have the exact same type.

[Ronan]

I don't have any other LM13X00's on hand at all, or even a stompbox that uses this chip so I can drop one into it.

When set up as a voltage controlled amplifier, like in the datasheet, it performed within specs, easily, although I was using rough and ready test gear, it would still have shown if the chip was not in spec. The only thing I didn't test was the linearizing input to the diodes. I was really impressed/surprised with how closely matched the 2 OTA's inside the IC were, in both samples I tested.

I'm assuming these IC's are innocent until proven guilty, I'll let you know if there are any issues.

[Ronan]

Since the last post, I built a phaser using 3 x LM13700's. I did have an issue with hiss with one IC, and replaced it and all good. Then I bought 3 x LM13700's for around $3.40 each from RS Components in Australia (free shipping). I fitted 2 of the RS Components IC's to this new phaser (I only socketed 2 of the 3 LM13700's - long story) and could not hear any difference in sound at all, real or imagined.

The IC's from RS had the same shape "N" symbol and were sold as Texas Instruments stock. The laser marking, including the etching of the "N" symbol, was not the same as the suspect ones. Also, the pins on the suspect units were not nearly as strong as the pins on most "normal" IC's, and appeared to have a fairly rough tinning on them. I always have to bend the pins to a right angle on a new IC, to fit it in the pcb or socket, that's how I know how strong the pins should be (resistant to bending).

My conclusion at this stage is, the suspect IC's are not the real thing, but can perform the same.

Check this thread, it may be possible that copied IC's are actually being produced? Is it possible?
http://www.diystompboxes.com/smfforum/index.php?topic=100004.msg878244#msg878244

[R.G.]

Check this thread, it may be possible that copied IC's are actually being produced? Is it possible?

It is 100% possible - and a known problem in the semiconductor industry.

[Toney]

 Would their really be a big enough marked to justify the tooling-up expenses for the fakers to manufacture these from scratch?
Both the PT2399 and the LM13xxx would only presumably be able to be sold to the hobby market, probably via Ebay and the Chinese equivalents. I just cant imagine a viable market for the prospective fake manufacturers.  I have both fake J201s and the same looking suss LM13600s as Ronan, so I know it happens, but there has to be more to the story. Sanded and branded would seem more likely.