Author Topic: Can somebody tell me, if that die is an LTZ1000?  (Read 1663 times)

0 Members and 1 Guest are viewing this topic.

Offline TryerTopic starter

  • Contributor
  • Posts: 16
  • Country: gr
Can somebody tell me, if that die is an LTZ1000?
« on: July 11, 2020, 02:08:22 pm »
 :-DD
I used to order electronic parts from China several years now. Yes, only because of the price. The delivery time varied from 3 to 5 weeks.
As i used them for my hobby only, i could live with it.
Last year, i ordered several reference ics to have some indication about my DMM.
Built a small portable reference with the AD588BQ for +5 and +10V and indeed it looked OK.
After some months and trying to improve this voltage reference with a temperature stabilizer,
i managed to kill the AD588. No problem, i ordered again 2 pieces from China.
Last week they were here. Replaced the damaged IC and switched on. Nothing.
Current was a too high (IC got warm). As i had taken all measures against electrostatism, i sweared a bit and took the 2nd one.
The same crap. I opened the 1st IC (which i had killed) and one of the second delivery. What the hell?... different dies... OK it was rhetorical question in the subject...
Yes, i heard about fakes but believed they would not dare with me.
Then my mind went to the 5xOPA2107AP i had received two weeks ago. Had already put them in a drawer for difficult times.... Had given also five stars...
Cheap, for 5 USD (including postage).
For a fast characterization i compared them to a TL082.
Connected one as a voltage follower with a 2MHz rectangular signal abt. 1Vpp, no offset.
You can see the attached results.
I never had that bad experiencies with parts from China, but now i have definetely begun rethinkig...

It is not about the money. It was refunded.
It is the waste of time. It is about being cheated.
It is like something defective that gets not repaired.

PS Sorry for the bad quality of the pics, my hands were trembling because of my anger ... |O
« Last Edit: July 11, 2020, 03:16:07 pm by Tryer »
 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4543
  • Country: ua
    • xDevs.com
Re: Can somebody tell me, if that die is an LTZ1000?
« Reply #1 on: July 11, 2020, 02:36:42 pm »
You get what you paid for, why anger then? It's nothing to do with LTZ either.
YouTube | Metrology IRC Chat room | Let's share T&M documentation? Upload! No upload limits for firmwares, photos, files.
 

Offline martinr33

  • Frequent Contributor
  • **
  • Posts: 363
  • Country: us
I guess there are four levels:

 - totally fake, don't do anything or are just the wrong part.
 - Fake, correct pins, but poor performance. Remarked from an inferior pat
 - Upmarked - factory rejects or low-grade parts, correct function, but not in spec as marked.
 - Remarked, used in-spec parts but with possible handling damage. Usually look better than new because of the marking.


The clue is in the date code. Remarkers don't usually print variable date codes. So if all the parts have the same date code, they are suspect.
 
The following users thanked this post: Tryer

Online PA0PBZ

  • Super Contributor
  • ***
  • Posts: 5300
  • Country: nl
Keyboard error: Press F1 to continue.
 
The following users thanked this post: Tryer

Offline magic

  • Super Contributor
  • ***
  • Posts: 7542
  • Country: pl
Re: Can somebody tell me, if that die is an LTZ1000?
« Reply #4 on: July 11, 2020, 06:48:19 pm »
The only surprising part of this story is the 12V/µs slew rate of your TL082.

It should be 0.3V/µs, if you know what I mean ;) ;D
 

Offline wraper

  • Supporter
  • ****
  • Posts: 18127
  • Country: lv
The clue is in the date code. Remarkers don't usually print variable date codes. So if all the parts have the same date code, they are suspect.
Nothing wrong with the same code as such. Actually when you get a bunch of parts with different date codes, they might be salvaged. The problem is when you see the same date code on parts with some package differences.
 

Offline magic

  • Super Contributor
  • ***
  • Posts: 7542
  • Country: pl
Re: Can somebody tell me, if that die is an LTZ1000?
« Reply #6 on: July 11, 2020, 06:59:06 pm »
If it costs 10% of retail price then yes, there is a big problem with same date codes.

Recycled is better than fake.
 
The following users thanked this post: Tryer

Offline TryerTopic starter

  • Contributor
  • Posts: 16
  • Country: gr
Re: Can somebody tell me, if that die is an LTZ1000?
« Reply #7 on: July 11, 2020, 07:58:21 pm »
Thanks a lot for all responses. I learned a lot.
@magic : the TL082 has a slew rate of 13V/us, acc. to the datasheet. I see 12.5V/us (abt. 1V/80ns) on the scope.
The steep curve is the input signal. So it is quite close to the spec value. I think they also use another test circuit with a certain R//C load.
I did not intent to measure acc. to spec. Just compared them to each other. And look at the offset.
The OPA2107 should be faster (18V/ us) but is much slower.
 

Offline magic

  • Super Contributor
  • ***
  • Posts: 7542
  • Country: pl
Re: Can somebody tell me, if that die is an LTZ1000?
« Reply #8 on: July 11, 2020, 09:51:12 pm »
I was kidding.

0.3V/µs is the spec of LM358, which is what you get when you buy TL082 for the price of LM358 on auction sites. See here. Not sure how many of the more expensive ones out there are authentic, because for that project I deliberately sought out the cheapest and shadiest crap.

Bottom line, lots of stuff on sites like AliBay is fake and it's been that way for many years. Consider yourself lucky getting your first authentic (probably recycled) AD588. It's a mix of lottery and skill to get anything usable from there. There are threads about other references from China too if you look around, with mixed bag results.

If you would like to try to identify the fake 2107, start a thread in some more general section.
 
The following users thanked this post: Tryer


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf