Prev: teleprompter malfunction
Next: Metcal: lost their minds
From: Joerg on 26 Jan 2010 19:14
Fred Bartoli wrote:
> John Larkin a �crit :
>> On Mon, 25 Jan 2010 23:37:18 +0100, Fred Bartoli <" "> wrote:
>>
>>> I'm searching some BZX884 2V4 & 3V6 NXP/Philips zeners for prototyping.
>>> These only sell by the 20k pieces wheel and I want to make some
>>> measurements before buying a wheel for maybe nothing.
>>>
>>> Anyone having a few of those to send?
>>>
>>> Thanks in advance.
>>
>> 2.4 volt zeners are pretty terrible regulators... look at their
>> dynamic impedance and TCs. But they are cheap.
>>
>
> I know that, and that's why I need some to measure and make a test. I'm
> not terribly exigent for the reg and the circuit will withstand that,
> but depending on it's real curves it might pose a startup problem I want
> to check. Fairchild has some interesting curves in
> http://www.fairchildsemi.com/ds/MM%2FMM5Z3V3.pdf
> but I've seen some big datasheet variations between brands on those low
> voltage diodes, much much more than with more reasonable voltages, so I
> can't take some other brand to check. I'm not even sure that just a
> different package, still from NXP, will use the same die. I'll check
> that with them.
>
> Why try to use it? Because it's the tiniest reference I've found, even
> if crappy. Only 0.6x1mm, with leads under the body. I'll have to cram 27
> components on a 7mm diameter pcb, plus 4 connect pads. (4 layers,
> microvia in pad)
> Some others are available in a frighteningly huge SOD523/SC79 0.9x1.7mm
> but I'm afraid I can't afford the 2.5 area.
>

If you can let off a little more about your circuit maybe there are
other options. For example the LT6700 comes in a 2mm by 3mm pack,
contact pads underneath, and in it you get two comparators plus a nice
400mV reference:

http://cds.linear.com/docs/Datasheet/6700123ff.pdf

But if you consider it write me an email. It has an IMHO nasty bug
during startup that might trip you up (it almost did in my case).


>> Can't you get samples from a rep?
>>
>
> I've checked the usual suspects and none have stock, so back to square one.
>
> Oh, and reps, here in France, are really unhelpful. Yeah, I'm a just a
> (no, not a gigolo) consultant (a pretty rare species here), and the
> customer will be delighted to use half a wheel a year, so nothing to
> dream about for the reps.
>

From past experience I can only tell you that measuring a few samples
guarantees nothing with low voltage zeners. The shipped reel might be
different. You really need to get the quality control limit sheets from
the manufacturer. Done it several times, they usually give that to you
but sometimes you have to sign an NDA. Otherwise you may end up with a
product that runs fine, suddenly your client sells more than
anticipated, another reel is purchased and ... boink ... game over. And
then they all look at Monsieur le consultant :-)

It is also not guaranteed that all the zeners on one reel are from the
same wafer or even the same production week.

--
Regards, Joerg

http://www.analogconsultants.com/

"gmail" domain blocked because of excessive spam.
Use another domain or send PM.
From: Jim Thompson on 26 Jan 2010 19:46
On Tue, 26 Jan 2010 16:14:45 -0800, Joerg <invalid(a)invalid.invalid>
wrote:

>Fred Bartoli wrote:
>> John Larkin a �crit :
>>> On Mon, 25 Jan 2010 23:37:18 +0100, Fred Bartoli <" "> wrote:
>>>
>>>> I'm searching some BZX884 2V4 & 3V6 NXP/Philips zeners for prototyping.
>>>> These only sell by the 20k pieces wheel and I want to make some
>>>> measurements before buying a wheel for maybe nothing.
>>>>
>>>> Anyone having a few of those to send?
>>>>
>>>> Thanks in advance.
>>>
>>> 2.4 volt zeners are pretty terrible regulators... look at their
>>> dynamic impedance and TCs. But they are cheap.
>>>
>>
>> I know that, and that's why I need some to measure and make a test. I'm
>> not terribly exigent for the reg and the circuit will withstand that,
>> but depending on it's real curves it might pose a startup problem I want
>> to check. Fairchild has some interesting curves in
>> http://www.fairchildsemi.com/ds/MM%2FMM5Z3V3.pdf
>> but I've seen some big datasheet variations between brands on those low
>> voltage diodes, much much more than with more reasonable voltages, so I
>> can't take some other brand to check. I'm not even sure that just a
>> different package, still from NXP, will use the same die. I'll check
>> that with them.
>>
>> Why try to use it? Because it's the tiniest reference I've found, even
>> if crappy. Only 0.6x1mm, with leads under the body. I'll have to cram 27
>> components on a 7mm diameter pcb, plus 4 connect pads. (4 layers,
>> microvia in pad)
>> Some others are available in a frighteningly huge SOD523/SC79 0.9x1.7mm
>> but I'm afraid I can't afford the 2.5 area.
>>
>
>If you can let off a little more about your circuit maybe there are
>other options. For example the LT6700 comes in a 2mm by 3mm pack,
>contact pads underneath, and in it you get two comparators plus a nice
>400mV reference:
>
>http://cds.linear.com/docs/Datasheet/6700123ff.pdf
>
>But if you consider it write me an email. It has an IMHO nasty bug
>during startup that might trip you up (it almost did in my case).
>
>
>>> Can't you get samples from a rep?
>>>
>>
>> I've checked the usual suspects and none have stock, so back to square one.
>>
>> Oh, and reps, here in France, are really unhelpful. Yeah, I'm a just a
>> (no, not a gigolo) consultant (a pretty rare species here), and the
>> customer will be delighted to use half a wheel a year, so nothing to
>> dream about for the reps.
>>
>
> From past experience I can only tell you that measuring a few samples
>guarantees nothing with low voltage zeners. The shipped reel might be
>different. You really need to get the quality control limit sheets from
>the manufacturer. Done it several times, they usually give that to you
>but sometimes you have to sign an NDA. Otherwise you may end up with a
>product that runs fine, suddenly your client sells more than
>anticipated, another reel is purchased and ... boink ... game over. And
>then they all look at Monsieur le consultant :-)
>
>It is also not guaranteed that all the zeners on one reel are from the
>same wafer or even the same production week.

Isn't it sort of idiotic to look at zeners below ~5V, when there's the
TL431, even CMOS versions?

...Jim Thompson
--
| James E.Thompson, CTO | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |

I love to cook with wine. Sometimes I even put it in the food.
From: Fred Bartoli on 26 Jan 2010 20:15
Jim Thompson a �crit :
> On Tue, 26 Jan 2010 16:14:45 -0800, Joerg <invalid(a)invalid.invalid>
> wrote:
>
>> Fred Bartoli wrote:
>>> John Larkin a �crit :
>>>> On Mon, 25 Jan 2010 23:37:18 +0100, Fred Bartoli <" "> wrote:
>>>>
>>>>> I'm searching some BZX884 2V4 & 3V6 NXP/Philips zeners for prototyping.
>>>>> These only sell by the 20k pieces wheel and I want to make some
>>>>> measurements before buying a wheel for maybe nothing.
>>>>>
>>>>> Anyone having a few of those to send?
>>>>>
>>>>> Thanks in advance.
>>>> 2.4 volt zeners are pretty terrible regulators... look at their
>>>> dynamic impedance and TCs. But they are cheap.
>>>>
>>> I know that, and that's why I need some to measure and make a test. I'm
>>> not terribly exigent for the reg and the circuit will withstand that,
>>> but depending on it's real curves it might pose a startup problem I want
>>> to check. Fairchild has some interesting curves in
>>> http://www.fairchildsemi.com/ds/MM%2FMM5Z3V3.pdf
>>> but I've seen some big datasheet variations between brands on those low
>>> voltage diodes, much much more than with more reasonable voltages, so I
>>> can't take some other brand to check. I'm not even sure that just a
>>> different package, still from NXP, will use the same die. I'll check
>>> that with them.
>>>
>>> Why try to use it? Because it's the tiniest reference I've found, even
>>> if crappy. Only 0.6x1mm, with leads under the body. I'll have to cram 27
>>> components on a 7mm diameter pcb, plus 4 connect pads. (4 layers,
>>> microvia in pad)
>>> Some others are available in a frighteningly huge SOD523/SC79 0.9x1.7mm
>>> but I'm afraid I can't afford the 2.5 area.
>>>
>> If you can let off a little more about your circuit maybe there are
>> other options. For example the LT6700 comes in a 2mm by 3mm pack,
>> contact pads underneath, and in it you get two comparators plus a nice
>> 400mV reference:
>>
>> http://cds.linear.com/docs/Datasheet/6700123ff.pdf
>>
>> But if you consider it write me an email. It has an IMHO nasty bug
>> during startup that might trip you up (it almost did in my case).
>>
>>
>>>> Can't you get samples from a rep?
>>>>
>>> I've checked the usual suspects and none have stock, so back to square one.
>>>
>>> Oh, and reps, here in France, are really unhelpful. Yeah, I'm a just a
>>> (no, not a gigolo) consultant (a pretty rare species here), and the
>>> customer will be delighted to use half a wheel a year, so nothing to
>>> dream about for the reps.
>>>
>> From past experience I can only tell you that measuring a few samples
>> guarantees nothing with low voltage zeners. The shipped reel might be
>> different. You really need to get the quality control limit sheets from
>> the manufacturer. Done it several times, they usually give that to you
>> but sometimes you have to sign an NDA. Otherwise you may end up with a
>> product that runs fine, suddenly your client sells more than
>> anticipated, another reel is purchased and ... boink ... game over. And
>> then they all look at Monsieur le consultant :-)
>>
>> It is also not guaranteed that all the zeners on one reel are from the
>> same wafer or even the same production week.
>
> Isn't it sort of idiotic to look at zeners below ~5V, when there's the
> TL431, even CMOS versions?
>
> ...Jim Thompson

Have one in under a 1mmx1mm area, pads included package?

I of course have looked that path. But you've made the die too big :-)

--
Thanks,
Fred.
From: Jim Thompson on 26 Jan 2010 20:33
On Wed, 27 Jan 2010 02:15:17 +0100, Fred Bartoli <" "> wrote:

>Jim Thompson a �crit :
>> On Tue, 26 Jan 2010 16:14:45 -0800, Joerg <invalid(a)invalid.invalid>
>> wrote:
>>
>>> Fred Bartoli wrote:
>>>> John Larkin a �crit :
>>>>> On Mon, 25 Jan 2010 23:37:18 +0100, Fred Bartoli <" "> wrote:
>>>>>
>>>>>> I'm searching some BZX884 2V4 & 3V6 NXP/Philips zeners for prototyping.
>>>>>> These only sell by the 20k pieces wheel and I want to make some
>>>>>> measurements before buying a wheel for maybe nothing.
>>>>>>
>>>>>> Anyone having a few of those to send?
>>>>>>
>>>>>> Thanks in advance.
>>>>> 2.4 volt zeners are pretty terrible regulators... look at their
>>>>> dynamic impedance and TCs. But they are cheap.
>>>>>
>>>> I know that, and that's why I need some to measure and make a test. I'm
>>>> not terribly exigent for the reg and the circuit will withstand that,
>>>> but depending on it's real curves it might pose a startup problem I want
>>>> to check. Fairchild has some interesting curves in
>>>> http://www.fairchildsemi.com/ds/MM%2FMM5Z3V3.pdf
>>>> but I've seen some big datasheet variations between brands on those low
>>>> voltage diodes, much much more than with more reasonable voltages, so I
>>>> can't take some other brand to check. I'm not even sure that just a
>>>> different package, still from NXP, will use the same die. I'll check
>>>> that with them.
>>>>
>>>> Why try to use it? Because it's the tiniest reference I've found, even
>>>> if crappy. Only 0.6x1mm, with leads under the body. I'll have to cram 27
>>>> components on a 7mm diameter pcb, plus 4 connect pads. (4 layers,
>>>> microvia in pad)
>>>> Some others are available in a frighteningly huge SOD523/SC79 0.9x1.7mm
>>>> but I'm afraid I can't afford the 2.5 area.
>>>>
>>> If you can let off a little more about your circuit maybe there are
>>> other options. For example the LT6700 comes in a 2mm by 3mm pack,
>>> contact pads underneath, and in it you get two comparators plus a nice
>>> 400mV reference:
>>>
>>> http://cds.linear.com/docs/Datasheet/6700123ff.pdf
>>>
>>> But if you consider it write me an email. It has an IMHO nasty bug
>>> during startup that might trip you up (it almost did in my case).
>>>
>>>
>>>>> Can't you get samples from a rep?
>>>>>
>>>> I've checked the usual suspects and none have stock, so back to square one.
>>>>
>>>> Oh, and reps, here in France, are really unhelpful. Yeah, I'm a just a
>>>> (no, not a gigolo) consultant (a pretty rare species here), and the
>>>> customer will be delighted to use half a wheel a year, so nothing to
>>>> dream about for the reps.
>>>>
>>> From past experience I can only tell you that measuring a few samples
>>> guarantees nothing with low voltage zeners. The shipped reel might be
>>> different. You really need to get the quality control limit sheets from
>>> the manufacturer. Done it several times, they usually give that to you
>>> but sometimes you have to sign an NDA. Otherwise you may end up with a
>>> product that runs fine, suddenly your client sells more than
>>> anticipated, another reel is purchased and ... boink ... game over. And
>>> then they all look at Monsieur le consultant :-)
>>>
>>> It is also not guaranteed that all the zeners on one reel are from the
>>> same wafer or even the same production week.
>>
>> Isn't it sort of idiotic to look at zeners below ~5V, when there's the
>> TL431, even CMOS versions?
>>
>> ...Jim Thompson
>
>Have one in under a 1mmx1mm area, pads included package?
>
>I of course have looked that path. But you've made the die too big :-)

I didn't design that particular circuit. It dates to the early '70's
when I was making switchers for GenRad. I design similar structures
into my custom chips... almost monthly ;-)

Have you pursued the CMOS version? It should be available in a small
package.

...Jim Thompson
--
| James E.Thompson, CTO | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |

I love to cook with wine. Sometimes I even put it in the food.
From: Fred Bartoli on 26 Jan 2010 20:57
Jim Thompson a �crit :
> On Wed, 27 Jan 2010 02:15:17 +0100, Fred Bartoli <" "> wrote:
>
>> Jim Thompson a �crit :
>>> On Tue, 26 Jan 2010 16:14:45 -0800, Joerg <invalid(a)invalid.invalid>
>>> wrote:
>>>
>>>> Fred Bartoli wrote:
>>>>> John Larkin a �crit :
>>>>>> On Mon, 25 Jan 2010 23:37:18 +0100, Fred Bartoli <" "> wrote:
>>>>>>
>>>>>>> I'm searching some BZX884 2V4 & 3V6 NXP/Philips zeners for prototyping.
>>>>>>> These only sell by the 20k pieces wheel and I want to make some
>>>>>>> measurements before buying a wheel for maybe nothing.
>>>>>>>
>>>>>>> Anyone having a few of those to send?
>>>>>>>
>>>>>>> Thanks in advance.
>>>>>> 2.4 volt zeners are pretty terrible regulators... look at their
>>>>>> dynamic impedance and TCs. But they are cheap.
>>>>>>
>>>>> I know that, and that's why I need some to measure and make a test. I'm
>>>>> not terribly exigent for the reg and the circuit will withstand that,
>>>>> but depending on it's real curves it might pose a startup problem I want
>>>>> to check. Fairchild has some interesting curves in
>>>>> http://www.fairchildsemi.com/ds/MM%2FMM5Z3V3.pdf
>>>>> but I've seen some big datasheet variations between brands on those low
>>>>> voltage diodes, much much more than with more reasonable voltages, so I
>>>>> can't take some other brand to check. I'm not even sure that just a
>>>>> different package, still from NXP, will use the same die. I'll check
>>>>> that with them.
>>>>>
>>>>> Why try to use it? Because it's the tiniest reference I've found, even
>>>>> if crappy. Only 0.6x1mm, with leads under the body. I'll have to cram 27
>>>>> components on a 7mm diameter pcb, plus 4 connect pads. (4 layers,
>>>>> microvia in pad)
>>>>> Some others are available in a frighteningly huge SOD523/SC79 0.9x1.7mm
>>>>> but I'm afraid I can't afford the 2.5 area.
>>>>>
>>>> If you can let off a little more about your circuit maybe there are
>>>> other options. For example the LT6700 comes in a 2mm by 3mm pack,
>>>> contact pads underneath, and in it you get two comparators plus a nice
>>>> 400mV reference:
>>>>
>>>> http://cds.linear.com/docs/Datasheet/6700123ff.pdf
>>>>
>>>> But if you consider it write me an email. It has an IMHO nasty bug
>>>> during startup that might trip you up (it almost did in my case).
>>>>
>>>>
>>>>>> Can't you get samples from a rep?
>>>>>>
>>>>> I've checked the usual suspects and none have stock, so back to square one.
>>>>>
>>>>> Oh, and reps, here in France, are really unhelpful. Yeah, I'm a just a
>>>>> (no, not a gigolo) consultant (a pretty rare species here), and the
>>>>> customer will be delighted to use half a wheel a year, so nothing to
>>>>> dream about for the reps.
>>>>>
>>>> From past experience I can only tell you that measuring a few samples
>>>> guarantees nothing with low voltage zeners. The shipped reel might be
>>>> different. You really need to get the quality control limit sheets from
>>>> the manufacturer. Done it several times, they usually give that to you
>>>> but sometimes you have to sign an NDA. Otherwise you may end up with a
>>>> product that runs fine, suddenly your client sells more than
>>>> anticipated, another reel is purchased and ... boink ... game over. And
>>>> then they all look at Monsieur le consultant :-)
>>>>
>>>> It is also not guaranteed that all the zeners on one reel are from the
>>>> same wafer or even the same production week.
>>> Isn't it sort of idiotic to look at zeners below ~5V, when there's the
>>> TL431, even CMOS versions?
>>>
>>> ...Jim Thompson
>> Have one in under a 1mmx1mm area, pads included package?
>>
>> I of course have looked that path. But you've made the die too big :-)
>
> I didn't design that particular circuit. It dates to the early '70's
> when I was making switchers for GenRad. I design similar structures
> into my custom chips... almost monthly ;-)
>
> Have you pursued the CMOS version? It should be available in a small
> package.
>

IIRC, I did look at some CMOS version, because I could have used it as
an opamp with offseted input, which is almost what I'm doing, but I
think I came with nothing.
He he... Have a reference?

--
Thanks,
Fred.
First  |  Prev  |  Next  |  Last
Pages: 1 2 3
Prev: teleprompter malfunction
Next: Metcal: lost their minds