This EL2018CN seems to have vanished from inventory of suppliers like Digikey as well as others. I can however find it on eBay. I’m in no need for this IC, just curious. Is it obsolete?
In my 3458A (probably manufactured 2011-2013 and bought new in Norway by a company in 2013; previous owner) there are two patch boards with EL2252CM. TiN has mentioned this patch board in “Repair of second HP 3458A metrology DMM”.
A2 (ERC 5314): Patch board replaces U501. There is also a strap that makes this patch board a little different from the patch board on A3. Because it’s replacing only one EL2018CN?
See picture 1.
A3 (ERC 5145): On this board there is a single EL2018CN, U181 and a patch board replacing U142 and U405.
See picture 2.
I wonder why they didn’t replace U181 with the same patch board as in A2?
Finally, I would again like to thank Dr. Frank, Kleinstein and others who contributed with their knowledge, expertise and suggestions to a successful repair of my 3458A in 2020.
Franklin
Finding the extra boards to replace the EL2018 is a clear indication the parts are obsolete and very hard to get even for HP/KS. So it is no longer available already for quite some time. There seem to be no direct replacement and likely also limited need. Chances it is no longer produced because of processes from the original source Elantec are no longer available. For the new black eddition they use a different comparator with extra +-5 V regulators and input protection.
U181 is an odd one. I still don't fully understand this part. From the circuit I see absolutely no need for this to be a super fast comparator: it compares ground to a signal from an LM358. A LM311 / LM393 should be suficient here, if this part is enabled at all. It could be enable funtion why this chip is used.
The patch boards may be different age and added when refurbishing the boards.
The China brokers still have some inventory. I've picked some up.
In most cases, they are obvious remarks. Pefect flat tops, beautiful white lettering - and incorrect logos and too-shallow notches from the grinding.
In other cases, they look legit but are clearly pulls, with straight legs and "socket scratches" on the metal.
Finally, there are few with original-looking markings and correctly angled legs.
Probably not a bad idea to repro the adapter boards.
On the apparently unnecessarily fast comparator at U181 - sometimes, it is better to use common parts across the design. It may be that it was just cheaper to use the Elantec part than another comparator.
Maybe it is possible to salvage U181 by replacing it with an available part - that is worth a look.
Bought 6 from a vendor in China, each cost around USD$2. NOS, no solder on the legs.
Did a side by side comparison on a A3 board U181, looks identical
The optical impression looks OK, but there are plenty of good looking fakes.
As these are DIP package, one could do a quick electrical test on a bread board. So is the supply current about right ?
Did a side by side comparison on a A3 board U181, looks identical
only on the first view.
a original manufacturer would never print the part type so near to the edge of the package.
The original part has the printing symmetrically.
Did you already make the acetone test?
I guess on the "identical" parts you will find that there is some "black topping" on the package.
Are the markings on the bottom side of the ICs identical?
with best regards
Andreas
Here's a picture of some Elantec parts that I have accumulated from several sources on eBay.
I think these are genuine. Consistent printing, no visible blacktopping, consistent molding marks on the rear (they all match the two I show upside down). The logos are also consistent, and look like the Elantec logo.
The pins are sometimes straight, sometimes at the correct angle - but the angled ones show clear signs of adjustment (occasional bent legs).
Given the date codes and leg cues, I think these are all genuine but recycled. Mixed date codes in the same tube of 10 are also a clue.
I agree on the suspect nature of the parts above. I'd like to see a bigger picture of the board to see if the date codes are consistent with the other chips.
No black topping found and the mark on the bottom of those chips are marked as “Malaysia”
Did some measurement with the supply current (Vs +- 15V)
with both input 1k ohm to ground:
IS+EN 8.79 ma (data sheet typical 8.4 max 12)
IS+DIS 4.62 ma (data sheet typical 4.7 max 6)
IS-EN 13.07 ma (data sheet typical 13.0 max 17)
IS-DIS 5.16 ma (data sheet typical 5.0 max 6.5)
Note: @martinr33 The photo on my previous post is means to show a side by side comparison with the one on the A3 board
The supply current looks good - so they really seem to the right chip. It could still be pulls with some new makeup, but for most used this would not matter. For an otherwise unobtanium chip a working recycled chip is still acceptable in most cases.
The supply current looks good - so they really seem to the right chip. It could still be pulls with some new makeup, but for most used this would not matter. For an otherwise unobtanium chip a working recycled chip is still acceptable in most cases.
I just bough a few more from another vendor who offer half the price
but say they are recycled. Will post update once I received those chips.
Batch #2, all pins has solder on it, likely recycled. Date code 94xx, manufactured in Philippine. Cost only USD $0.6
Did some measurement with the supply current (Vs +- 15V)
with both input 1k ohm to ground:
IS+EN 8.19 ma (data sheet typical 8.4 max 12)
IS+DIS 4.28 ma (data sheet typical 4.7 max 6)
IS-EN 12.10 ma (data sheet typical 13.0 max 17)
IS-DIS 5.00 ma (data sheet typical 5.0 max 6.5)
Batch #3, all pins has NO solder on it, NOS ? Date code 9609, manufactured in Malaysia. Cost USD $1.53
Did some measurement with the supply current (Vs +- 15V)
with both input 1k ohm to ground:
IS+EN 4.98 ma (data sheet typical 8.4 max 12)
IS+DIS 2.27 ma (data sheet typical 4.7 max 6)
IS-EN 9.54 ma (data sheet typical 13.0 max 17)
IS-DIS 3.23 ma (data sheet typical 5.0 max 6.5)
Fake
If so, wonder what is inside as it response to the voltage level change on CS.
The measured supply currents neither violate typical (typical means typical) nor max current specs. No conclusion.
You need to setup a test circuit for a functional check. Put a sine wave on one of the inputs and another digital signal onto the enable and take a record with a scope to see whether the IC works. If it works, the next step is a characterization with a small input signal of high frequency and with various common mode input voltages.
Regards, Dieter
Just for reference:
3458A Worklog: AC error fixed - EL2039 replacementTop left is faulty specimen, bottom left replacement spare (other measured fine and already soldered in), right are fakes - as different packages and finally acetone test revealed:
