EMRFD Message Archive 8687

Message Date From Subject
8687 2013-05-24 20:41:40 AD7ZU precision parts
 
 
I had a question before committing to ordering a pile of parts.  One of my coworkers informed me that common passive parts i.e. resistors, caps etc are routinely prescreened to extract the tight tolerance parts by manufacturers.  Say I need some .01uf caps to 1%.  I plan on ordering 200 20% parts and then hand selecting to find 10 that are within 1%.  Is it true that there would likely be no parts within 1% in the 20% 200 part bag?
And that if I want 10 resistors within 0.1% I would just have to order the precision parts?  since the manufacturer would have already screened the close value parts from the 10% bin?
I just wondered if anyone had any experience with attempting to hand select a few parts from a low tolerance batch?  There is a huge premium on precision tolerance parts.
 
Thanks,
Randy
AD7ZU

[Non-text portions of this message have been removed]
8689 2013-05-24 22:04:13 Andy Re: precision parts
AD7ZU <ad7zu@yahoo.com> wrote:


> I had a question before committing to ordering a pile of parts. One of my
> coworkers informed me that common passive parts i.e. resistors, caps etc
> are routinely prescreened to extract the tight tolerance parts by
> manufacturers. Say I need some .01uf caps to 1%. I plan on ordering 200
> 20% parts and then hand selecting to find 10 that are within 1%. Is it
> true that there would likely be no parts within 1% in the 20% 200 part bag?
>

Yes ... sometimes that is true. In the old days it was true of resistors.
If you ordered 20% resistors, half of them would be 10-20% high and the
other half was 10-20% low. Bathtub or double hump distribution. That was
how they make and sell parts ... they test them; weed out the ones that are
way out of tolerance; pull out the ones that measure best and mark them
accordingly; etc.; and mark the rest as wider tolerance parts. It's not
like they have parallel manufacturing lines, one making 1% parts and the
other churning out 20% parts but constructed the same. One process line
makes them all. It's just a matter of "binning" them.

I have since been told that you might not get this happening anymore; but
I'm willing to bet it depends on the type of part and on the market. I
could envision it going either way.

You could get a situation with capacitors, where the 1% parts and 20% parts
use completely different chemistries (like ceramic vs. polystyrene, or Y5V
vs. X7R vs. NPO ceramic). Or where the 20% parts come from a different
vendor who has inferior quality controls. If it were me, personally I
might not want to use a 20% capacitor that was hand-picked, where I needed
a 1% capacitor. I'd be afraid it might have a drift problem or tempco or
voltage dependency.

Andy


[Non-text portions of this message have been removed]
8690 2013-05-25 08:24:05 Roelof Bakker Re: precision parts
Hello Randy,

As 1% metal film resistors are cheap these days, this is what I use most
of the time.
A while ago, I have been checking a bunch of 100 and found that many were
within the 0.1% from each other.
If you don't need an exact value, but some closely matched resistors, this
is easy accomplished.

However take in account that long term and temperature stability might be
an issue.

73,
Roelof Bakker, pa0rdt
8691 2013-05-27 05:43:19 ha5rxz Re: precision parts
Precisi
8692 2013-05-27 10:10:00 Paul Anderson Re: precision parts
I imagine it would be tempting to just add trimmer caps so you can dial it
in.

On Monday, May 27, 2013, ha5rxz wrote:

> Precisi
8693 2013-05-27 16:45:27 Ronald RiemVis Re: precision parts
Randy,

I live in China where electronics are very cheap.
Can I help you?

Tell me what you need and I can arrange.

Greetings,

Ronald


On 25 May 2013 03:52, AD7ZU <ad7zu@yahoo.com> wrote:

> **
>
>
>
>
> I had a question before committing to ordering a pile of parts. One of my
> coworkers informed me that common passive parts i.e. resistors, caps etc
> are routinely prescreened to extract the tight tolerance parts by
> manufacturers. Say I need some .01uf caps to 1%. I plan on ordering 200
> 20% parts and then hand selecting to find 10 that are within 1%. Is it
> true that there would likely be no parts within 1% in the 20% 200 part bag?
> And that if I want 10 resistors within 0.1% I would just have to order the
> precision parts? since the manufacturer would have already screened the
> close value parts from the 10% bin?
> I just wondered if anyone had any experience with attempting to hand
> select a few parts from a low tolerance batch? There is a huge premium on
> precision tolerance parts.
>
> Thanks,
> Randy
> AD7ZU
>
> [Non-text portions of this message have been removed]
>
>
>


[Non-text portions of this message have been removed]
8694 2013-05-27 16:52:46 dan edwards Re: precision parts
high voltage, hi current p.s. transformers are getting crazy pricing in usa, any more..i have  a 'bud' in VK who is sourcing some nice, hi-efficiency, toroidal units from somewhere in  BY land....any tips ?   73, w5xz, dan..


8695 2013-05-28 01:00:13 ha5rxz Re: precision parts
Let me know when you find some 0.2uF trimmers and I'll send a truck around to collect them :)

Peter HA5RXZ

8696 2013-05-29 09:46:57 Tim Re: precision parts
There is very very little need for 1% precision 0.01uF parts. What you probably want are "1% matched" parts for, e.g. I-Q audio frequency phasing networks.

If you need matched-to-1-percent 0.01uF parts, buying a bag of looser 5% or 10% tolerance parts and binning them yourself, is very very practical. At which point, it doesn't matter that none of them are within 1% of 0.01uF, but you will find lots of matched 0.0097uF parts and lots of e.g. matched 0.0103uF parts.

Back in the 1970's, 20% tolerance resistors were common, but today I think we'd be hard pressed to find anyone selling anything worse than 5% tolerance resistors.

Matching and binning inductors and capacitors typically used in AF phasing networks, can be done very practically with audio-frequency bridges (something lightly touched up
8697 2013-05-29 10:42:52 William Carver Re: precision parts
When I was an undergrad in college, circa 1965, one of my EE profs, a
very practical/intelligent/knowledgeable guy, told me that at one time
he found that performance of a consumer design he was working on was all
over the map and he found the composition resistors were always beyond
either 5% high or 5% low. The 5% resistors had been selected out! That
corresponds to the story your coworker tells.

But that was a long, long time ago. My experience with ordinary metal
film and SMD resistors have no such gap in them and they are generally
much better than 5%, I find many are within 1%. But 0.1% is a band 2%
wide within the 5% tolerance band. You'd need to select from a large
batch to find ten 0.1% tolerance.

>From a batch of 1% your odds are a lot better. Consider Vishay CMF55 1%
resistors: my Mouser catalog isn't current, but it shows $12.80 for 100
of them. They are rated for 50 ppm/C stability. the usual for "RN55C"
style 1% metal film resistors. I'm betting you'd easily find ten within
0.1%.

But consider Vishay PTF56, 1/8W, 0.1%: they are not available in very
many values, but if 4.99K, 10K, 20K, 100K kind of increments work for
you, they are $1 to $1.40 apiece so for ten pieces its kind of a wash
with selecting from 100 1%, except you'd have ninety resistors left over
for your junkbox, hi. But the 0.1% tolerance are rated for 10 ppm/C
stability, inherently a more stable part.

If temperature coefficient is not an issue, Vishay also has resistors
rated at 100 ppm/C, corresponding to "RN55D" spec. For normal values 10
ohms to 2.5 megohms they only cost $4.50 per hundred. This tilts you
back toward the selecti
8698 2013-05-29 17:10:52 KK7B Re: precision parts
Hi Bill,

Nice to see you at Dayton, though we didn't get a chance to talk much. My experience with modern 1% resistors is that they are pretty tight. It's easy to get ten within 0.1% within a few minutes of measuring. Since my phasing rigs use a lot of 10k resistors, I typically take a bag of 200, sit down with my Fluke 89 ohmmeter and a prototyping board, and start measuring. I put the low values at one end of the board and the high values at the other end, and leave them stored on the board. Then when I need a batch of 0.1% tolerance resistors, I just grab them off the board.

Best Regards,

Rick

8699 2013-05-29 18:55:50 William Carver Re: precision parts
I agree Rick, it's relatively easy with a good digital multimeter to get
well matched resistors. And with an AADE LC Meter, capacitors too. You
DO need to have a batch to select
8701 2013-05-29 19:41:50 AD7ZU Re: precision parts
 
I spent most of a career specifying standard bypass caps and 10k pullups with little regard for anything except cost and availability. Its a new awakening to be designing anything analog... I will just order several lots of 200pcs and sort and select R's and C's as needed.  The lower tolerance pcs will not go to waste.  these days i prefer  1206s ..easy to see, easy to solder, fewer holes in the pcb, and cheap.
 
As Bill notes today's parts are a bargin.
 
 
Thanks,
Randy
AD7ZU
 

________________________________
8704 2013-05-31 07:58:02 Jim Re: precision parts
And conveniently, if you build with DIP ICs, dead-bug style, a 1206 SMD will fit perfectly on two adjacent pins.

73
Jim N6OTQ



>________________________________
>