Source for low current zener in China?

[Joerg]

Hello Folks,

Had a tolerance issue with MMSZ4690 zeners and the manufacturing plant
in China said they were quoted 8 weeks for replacement reels. Of course,
that won't work.

Does anyone know a brand that is easily obtained in China and offers low
current zeners? What I am looking for is a 5.6V version in SOD-123
package that likes running at 200uA. I am going to look for US sources
now but that would mean several extra days to get them there.

 

[Winfield Hill]

Joerg wrote...

Hello Folks,

Had a tolerance issue with MMSZ4690 zeners and the manufacturing
plant in China said they were quoted 8 weeks for replacement
reels. Of course, that won't work.

Does anyone know a brand that is easily obtained in China and
offers low current zeners? What I am looking for is a 5.6V version
in SOD-123 package that likes running at 200uA. I am going to look
for US sources now but that would mean several extra days to get
them there.

Well, Joerg, it looks like you are in a pickle! You mention low
zener currents and a tolerance problem. Those two don't generally
go well together with zener diodes, which are commonly specified
at 20mA. One thing working in your favor, zener diodes in the
5- to 6-volt region have a pretty low zener impedance, which is
another way of saying their breakdown voltage doesn't change much
with current.

But you are having trouble with zeners voltage rated at 50uA,
when used at 200uA. Whew! Just how tight is your spec?

In such a situation, with cheap parts like these, I'd acquire a
reel or more and measure a few parts, before throwing the rest
into the manufacturing ring. Newark has 15k in stock, $0.045.

 

[Eeyore]

Hello Folks,

Had a tolerance issue with MMSZ4690 zeners and the manufacturing plant
in China said they were quoted 8 weeks for replacement reels. Of course,
that won't work.

Does anyone know a brand that is easily obtained in China and offers low
current zeners? What I am looking for is a 5.6V version in SOD-123
package that likes running at 200uA. I am going to look for US sources
now but that would mean several extra days to get them there.

I'm surprised.

I thought you only used jellybean parts !

Graham

 

[Eeyore]

How expensive are the CMOS equivalents of the venerable TL431?

A bit late for that I reckon.

Graham

 

[Jim Thompson]

Joerg wrote...

Well, Joerg, it looks like you are in a pickle! You mention low
zener currents and a tolerance problem. Those two don't generally
go well together with zener diodes, which are commonly specified
at 20mA. One thing working in your favor, zener diodes in the
5- to 6-volt region have a pretty low zener impedance, which is
another way of saying their breakdown voltage doesn't change much
with current.

But you are having trouble with zeners voltage rated at 50uA,
when used at 200uA. Whew! Just how tight is your spec?

In such a situation, with cheap parts like these, I'd acquire a
reel or more and measure a few parts, before throwing the rest
into the manufacturing ring. Newark has 15k in stock, $0.045.

How expensive are the CMOS equivalents of the venerable TL431?

...Jim Thompson

 

[Robert Baer]

Hello Folks,

Had a tolerance issue with MMSZ4690 zeners and the manufacturing plant
in China said they were quoted 8 weeks for replacement reels. Of course,
that won't work.

Does anyone know a brand that is easily obtained in China and offers low
current zeners? What I am looking for is a 5.6V version in SOD-123
package that likes running at 200uA. I am going to look for US sources
now but that would mean several extra days to get them there.

Central Semi has a few lines of low current spec zeners.

 

[Fred Bartoli]

Eeyore a écrit :

I'm surprised.

I thought you only used jellybean parts !

Not jellybean, but swallow nested parts here.

 

[Joerg]

Hello Win,

Well, Joerg, it looks like you are in a pickle! You mention low
zener currents and a tolerance problem. Those two don't generally
go well together with zener diodes, which are commonly specified
at 20mA. One thing working in your favor, zener diodes in the
5- to 6-volt region have a pretty low zener impedance, which is
another way of saying their breakdown voltage doesn't change much
with current.

But you are having trouble with zeners voltage rated at 50uA,
when used at 200uA. Whew! Just how tight is your spec?

Not very tight. The cut-off at end test is 5.1V which is well below the
data sheet tolerance band and that band was stated for 50uA. So
something seems to be wrong with the lot. I have two inquiries pending.
One about whether there is an epidemic problem or whether we can
scrounge reels from other places. Or find out which lots would be ok.
The other into Vishay which makes almost the same part but where there
is a data sheet discrepancy. Figure 7 in there shows the distinct <2mA
break-away of higher current zeners but the guaranteed data says the
contrary:

http://www.vishay.com/docs/85773/mmsz4681.pdf

Didn't get an answer yet and the clock is ticking :-(

In such a situation, with cheap parts like these, I'd acquire a
reel or more and measure a few parts, before throwing the rest
into the manufacturing ring. Newark has 15k in stock, $0.045.

Yes, that's what we are doing in parallel.

 

[Joerg]

Hello Jim,

How expensive are the CMOS equivalents of the venerable TL431?

Did that on another design for that client. You have to use the TLV431
because of the low cathode current, plus it adds a couple resistors so
it's increasing the cost by a few cents. But on this legacy design from
before my time with them they chose to keep the zeners. Which were fine
for many years but now they seem to be out of spec, big time. Just
baffles me how they could have passed QC in the first place.

 

[Joerg]

Hello Graham,

I'm surprised.

I thought you only used jellybean parts !

That part of the design is legacy, from before I got involved. But I'd
say zeners are almost jelly-bean parts these days.

 

[Joerg]

Hello Robert,

Central Semi has a few lines of low current spec zeners.

Yes, they do. CMHZ4626, but non-stock everywhere I checked. I found some
others as well but lead times are as high as 18wks and we need them
pretty much right now :-(

 

[Eeyore]

Hello Graham,


That part of the design is legacy, from before I got involved. But I'd
say zeners are almost jelly-bean parts these days.

It was the low current aspect that I thought was slightly unusual.

Graham

 

[Joerg]

Hello Graham,

It was the low current aspect that I thought was slightly unusual.

Not really these days. They may cost a cent more than regular ones but
on battery powered units you don't have a choice, except going to the
TLV431 which adds even more cost.

 

[Joerg]

Hello Spehro,

Could be a result of counterfeit parts.

Could be. However, this manufacturing plant is very professional in
their purchasing and I've got photos of the reels. Looks like the normal
stuff from ON's Malaysia plant.

 

[Joerg]

Hello Spehro,

There are some Japanese parts that should be quite available in China,
and are spec'd around 3% at 500uA, so they should be pretty tight at
0.2mA, but I'm not sure about that huge package of yours.

I checked the usual suspects but only found SOT23 and other (newer)
types of packages. This is a legacy design.

 

[Spehro Pefhany]

Joerg wrote...

Well, Joerg, it looks like you are in a pickle! You mention low
zener currents and a tolerance problem. Those two don't generally
go well together with zener diodes, which are commonly specified
at 20mA. One thing working in your favor, zener diodes in the
5- to 6-volt region have a pretty low zener impedance, which is
another way of saying their breakdown voltage doesn't change much
with current.

But you are having trouble with zeners voltage rated at 50uA,
when used at 200uA. Whew! Just how tight is your spec?

In such a situation, with cheap parts like these, I'd acquire a
reel or more and measure a few parts, before throwing the rest
into the manufacturing ring. Newark has 15k in stock, $0.045.

There are some Japanese parts that should be quite available in China,
and are spec'd around 3% at 500uA, so they should be pretty tight at
0.2mA, but I'm not sure about that huge package of yours.


Best regards,
Spehro Pefhany

 

[Spehro Pefhany]

data sheet tolerance band and that band was stated for 50uA. So
something seems to be wrong with the lot. I have two inquiries pending.
One about whether there is an epidemic problem or whether we can
scrounge reels from other places. Or find out which lots would be ok.
The other into Vishay which makes almost the same part but where there
is a data sheet discrepancy. Figure 7 in there shows the distinct <2mA
break-away of higher current zeners but the guaranteed data says the
contrary:

http://www.vishay.com/docs/85773/mmsz4681.pdf

Didn't get an answer yet and the clock is ticking :-(

Could be a result of counterfeit parts.


Best regards,
Spehro Pefhany

 

[Joerg]

Hello Spehro,

There are some in 0.8 x 2.1mm size, but I doubt the leads would reach
to your pads.

That would be a problem since this is automated assembly. I really was
hoping I had an answer from some of the mfgs by now but it seems the
care for existing customers (with substantial sales volume) isn't what
it used to be. Where are the good old days when it was still Motorola?

 

[Spehro Pefhany]

Hello Spehro,


I checked the usual suspects but only found SOT23 and other (newer)
types of packages. This is a legacy design.

There are some in 0.8 x 2.1mm size, but I doubt the leads would reach
to your pads.


Best regards,
Spehro Pefhany

 

[Frank Bemelman]

Hello Spehro,


That would be a problem since this is automated assembly. I really was
hoping I had an answer from some of the mfgs by now but it seems the care
for existing customers (with substantial sales volume) isn't what it used
to be. Where are the good old days when it was still Motorola?

No care for the existing and substantial customers, no care for the new
and small customers... the world is coming to an end.

The bottom line of the delay you experience is that they don't have a
solution for you. A "Yes-no problem" answer always arrives quick.
For that "No-we're sorry" answer, you always have to wait a bit longer.

Do we have a plan B ?