Author Topic: AD587LQ (Read 15924 times)

eurofox · « **on:** April 19, 2018, 02:38:46 pm »

Normally not available anymore

This ic was used in the venerable Geller lab SVR-T in ceramic version.

But I order 5 IC's from Aliexpress, I don't no if they are new, used, fake but I'll take the risk since it is cheap

eurofox

cellularmitosis · « **Reply #1 on:** April 19, 2018, 03:31:06 pm »

I took a similar risk on a few LQ's from taobao. I plugged them into my old Geller Labs reference board and they seemed to be wandering around. I intend to take a closer look at them soon but it appears they could be fake or have been damaged.

However, the AD586LQ's I ordered don't seem to be wandering, so maybe it wasn't totally a bust?

Good luck!

Andreas · « **Reply #2 on:** April 19, 2018, 06:52:48 pm »

Quote from: eurofox on April 19, 2018, 02:38:46 pm

but I'll take the risk since it is cheap

I hope it is worth the effort that you need to distinguish them from being genuine.
Some tips for quick checks:
- test the 1/f noise (perhaps they have sold a bandgap reference which has a factor 3-5 more noise).
- test the pin voltages (especially pin 8 the internal divided reference voltage and pin 3 which is usually the temperature output pin on cheaper band gaps).

I personally do not believe that you can get genuine unobtanium devices.

with best regards

Andreas

eurofox · « **Reply #3 on:** May 23, 2018, 02:34:08 pm »

Got from Aliexpress the message

I think the guy who try to sell this did not finally manage to get some of those litlle bad boy's

But I found a company that can supply them, after 4 weeks delay .......

I will check them, they look genuine and it they are good I will post the place

Kosmic · « **Reply #4 on:** May 23, 2018, 03:45:07 pm »

I'm looking at the specs and I don't see a lot of difference between AD587L and AD587U.

If the L version is not available anymore why not simply use the U version ?

eurofox · « **Reply #5 on:** May 23, 2018, 03:52:48 pm »

Quote from: Kosmic on May 23, 2018, 03:45:07 pm

I'm looking at the specs and I don't see a lot of difference between AD587L and AD587U.

If the L version is not available anymore why not simply use the U version ?

You should read this https://www.eevblog.com/forum/testgear/geller-labs-svr-and-svr-t/50/

Kosmic · « **Reply #6 on:** May 23, 2018, 04:11:00 pm »

Quote from: eurofox on May 23, 2018, 03:52:48 pm

Quote from: Kosmic on May 23, 2018, 03:45:07 pm
I'm looking at the specs and I don't see a lot of difference between AD587L and AD587U.

If the L version is not available anymore why not simply use the U version ?

You should read this https://www.eevblog.com/forum/testgear/geller-labs-svr-and-svr-t/50/

Ho ok, I understand now

GigaJoe · « **Reply #7 on:** May 23, 2018, 08:51:17 pm »

seen some for $20 ... kinda expensive.

is my understanding correct, a letter just a thermal instability, or something more? (excluding items with wide range of temperature where manufacturing process may differ) means it just more effort for thermo compensating circuit?

Andreas · « **Reply #8 on:** May 23, 2018, 09:40:24 pm »

Hello,

its either better selection for T.C. or better trimming.
But it is always a "average T.C." for the whole specified temperature range.
(measured at 3 temperatures: min, 25 deg C, and max of specified range e.g. 0..70 deg C or -55..125 deg C)

So if you use the device near room temperature the letter (grade) may be not helpful in all cases.

with best regards

Andreas

GigaJoe · « **Reply #9 on:** July 17, 2018, 03:18:23 am »

https://www.aliexpress.com/item/-/32881073708.html
got 6 reclaimed, mfg year 94, 95,91,04. all good, "philippines" on the back ...
drifting, stability etc - unknown ...

UPD: Seller started to ship relabeling chips ....

MiDi · « **Reply #10 on:** September 26, 2018, 08:23:06 pm »

Quote from: eurofox on May 23, 2018, 02:34:08 pm

Got from Aliexpress the message
I think the guy who try to sell this did not finally manage to get some of those litlle bad boy's

But I found a company that can supply them, after 4 weeks delay .......

I will check them, they look genuine and it they are good I will post the place

Any results so far?

Andreas · « **Reply #11 on:** September 26, 2018, 09:58:35 pm »

Hello,

Quote from: MiDi on September 26, 2018, 08:23:06 pm

Any results so far?

My preliminary results:

I have ordered 2 times (8 pcs each) from GigaJoe s seller.
https://www.eevblog.com/forum/metrology/ad587lq-from-aliexpress/msg1677863/#msg1677863

He needed both times 1-2 weeks until I could see in the tracking system that he had sent them out.
When I asked him for delivery he answered something like he wanted to find the best source for me.
So obviously the seller does not have them on stock: so quality may vary.

Each reference has a different date code.
Devices are desoldered with partly bent pins.
Output voltage on all 16 devices is within +/-3 mV from 10V. So even with ageing within 5mV datasheet spec.
Pin 3 shows 0 V (so no temperature output signal like on some band-gaps)
Pin 8 and pin 5 are around 7V so obviously it is a buried zener reference.
1/f noise is around 4uVpp.

up to now 2 devices have frequently (every few seconds) popcorn noise. (but have not evaluated all data up to now).
It seems that the popcorn candidates had been marked with a black dot.
On the first batch I thought that it was some flux residue from desoldering so I had removed it.
On the 2nd batch I can correlate the marked device with the popcorn noise.

T.C. is still under evaluation.
Measured 12 devices up to now with < 2.5 ppm/K over +/-15 deg C span around room temperature.
2 devices showed up to 2-4 ppm (ageing) drift during T.C. measurement.

with best regards

Andreas

GigaJoe · « **Reply #12 on:** September 26, 2018, 11:24:09 pm »

so, not bad in overall ...

my batch was send immediately ... ; no any labels
haven't able to test it,

one chip has strange output on pin 8 , like 1.$&% and up to 5V if heat it up ... aside of that pretty normal ...
and 1 completely fried. I can not tell it from this batch or another

JQ no LQ as described:
https://www.aliexpress.com/item/-/32832885977.html

plastic mostly fake .. don't even look at ... they actually do 10V output but nothing on pin-8 (even TC reasonable)

MiDi · « **Reply #13 on:** September 27, 2018, 06:13:20 am »

Andreas, you now have new nickname (at least for me): the machine

MiDi · « **Reply #14 on:** October 21, 2018, 01:10:03 pm »

10 pcs arrived yesterday from GigaJoe:

branadic · « **Reply #15 on:** October 21, 2018, 02:13:32 pm »

Looks like reference from 9540 has a crack?

-branadic-

MiDi · « **Reply #16 on:** October 21, 2018, 02:42:02 pm »

No, it is something like epoxy, hard and very sticky - could not scratch it away with fingernails.

Andreas · « **Reply #17 on:** October 21, 2018, 05:35:41 pm »

Quote from: Andreas on September 26, 2018, 09:58:35 pm

2 devices showed up to 2-4 ppm (ageing) drift during T.C. measurement.

Update.

I examined one of the drifters more in detail.
Result: still about 1-2 ppm drift during one temperature cycle (15 hours).
So I think the two "drifters" will not be usable for my 10V references.

with best regards

Andreas

MiDi · « **Reply #18 on:** October 23, 2018, 06:30:27 pm »

6 came DOA, 4 are very good within specs (worst is +2mV, best -0.5mV).
Seller refunded 6 pcs. immediatly.

syau · « **Reply #19 on:** October 24, 2018, 09:55:31 am »

Received my batch of shipment this Monday, 3 out of 10 DOA. The remaining 7 all within specification which the best one is only +2ppm

Macbeth · « **Reply #20 on:** November 10, 2018, 02:58:03 pm »

Ordered 6 from GigaJoes seller on 21/10 and received today. One is a dud, of the rest worst is +1.9mV, best +0.4mV. All CERDIP random dates.

WillTurner · « **Reply #21 on:** November 23, 2018, 05:08:01 am »

Voltage measurements of six parts from vendor LING LI "Shop1476012 Store" tested 28Aug18.

Datecode Vtrim Vout Notes
9144 7.22 10.01
9331 7.04 14.16 Bad
9816 13.91 14.34 Bad
9825 6 to 6.5 10.01 Vtrim unstable. Bad
0016 1.5 14.46 Bad
0422 7.06 10.01

Hence, at least four bad out of six parts. Note that I am not implying in any way that the remaining two parts are "good"

.

I sent the above information to the vendor, and their response was "Sorry, if the test does not pass, you should consider buying, because these ICs have been discontinued for a long time, and the market is also out of stock."

Based on the above, I could not recommend buying these. The AD587JQ through a trusted supply chain, and selection for tempco seems a better bet.

Andreas · « **Reply #22 on:** November 23, 2018, 08:36:56 pm »

Hello,

with 20 brand new parts from DigiKey AD587JQ I had only 3 parts
with less than 2 ppm/K and less than 5(6)uVpp popcorn noise.
(ok 5 other parts were not far away but the rest completely useless for my needs).
And even then you can have large drift (more than 2 ppm/year after 5 kHrs)
after pre-selection.

At least those parts seem to be only mis-used and recycled parts. (and no fakes).
I already had some parts labeled MAX6350 in CERDIP which were definitively no buried zeners.
So buying such parts from a non official source is always a bit of gambling.
And in every case you need the equipment for noise and drift tests.

With best regards

Andreas

GigaJoe · « **Reply #23 on:** November 24, 2018, 06:18:10 pm »

Does only AD587 has PIN-8 as an output from zener ? and no others in CDIP8 has it ? Reason why I'm asking, got relabeled AD587, but I'm suspicious it really 587 just a letters are relabeled.

last year bought got JQ (as pictured), not LQ ; seller are reluctant to talk ...
https://www.aliexpress.com/item/-/32832885977.html

Andreas · « **Reply #24 on:** November 24, 2018, 06:47:44 pm »

Hello,

Ref102 also has pin 8 as noise reduction input. (SO-8)
VRE310 also noise reduction on Pin 8.
Some other (5V) references on Pin 3.

What makes you shure that there are no re-packaged components.
From SO8 to CERDIP?

I always check Pin 3, Pin 5 and Pin 8 for suspicious voltages on unknown references.

with best regards

Andreas


EEVblog Main Site	EEVblog on Youtube	EEVblog on Twitter	EEVblog on Facebook	EEVblog on Odysee

Author Topic: AD587LQ (Read 15924 times)

Share me