Sourcing 0.1% resistors.

The only easily available 0.1% through hole resistors in all the 1% E96 values I've found are the IRC brand available at Mouser. Are there any other choices available in small quantities? Nothing wrong with the IRC's. I'm just curious.

What about this ?

http://fr.mouser.com/Passive-Components/Resistors/Through-Hole-Resistors/Metal-Film-Resistors/_/N-7h7zb?P=1z0vod7Z1z0vpm5Z1z0vl8vZ1z1414h&Keyword=Vishay&FS=True

They don't come in all E96 values. I'm making large rotary attenuators and I need lots of values.

Oh yes for that not enough true, hmm maybe some other company could have same ones with more Values???

PS, but you could still match most of the E96, if you buy a reel of lets say 50 and if they are of the same reel, they will be close enough for 50% even more, IIRC mines bought from Farnell, Mouser don't do 50 piece Reel i think, maybe RS also???, Just a tought...

I would need to have about 800 matched resistors... This is when the expensive stuff looks cheap.

If the values can be give or take 1% from the prescribed value, I found it easy to match resistors to .01%. I made matched stepped switches for a pultec and I only had to measure maybe 30 to find 4 of the same value. Some where off from the desired value by around 1% but as long as both channels were the same it worked for me.

I would need to have about 800 matched resistors... This is when the expensive stuff looks cheap.

Click to expand...

Sontec?

benlindell said:

If the values can be give or take 1% from the prescribed value

Click to expand...

This is for line amps and I want the absolute value to be the same for all eight attenuators.

You used to be able to get Holco but no more. All other brands I've seen need large orders or don't come in enough values. I guess IRC is the only option.

They are for 45 position two channel bridged T attenuators.  I already tried using .1% resistors in a pot style attenuators and the precision wasn't good enough. I want .5 dB steps with .1dB tolerance.

BTW, never tought of some SMD resistors? Well at least if you can solder them on the tabs? I don't know what style of attenuator you'll use, but for ex, those 24 stepped Uraltone the cheap chinese Elma's, you can solder SMD resistors on them, doing a bridge from one tab to another, need some seady hand tough, but i saw a pic on Ebay with pots sold like this, looks nice and clean, and you'll maybe find all the values you need? Just a tought.

I don't know if these are E96 values or not, but probably worth a look

http://www.rapidonline.com/Electronic-Components/Resistors-Potentiometer/Metal-Film-Resistors/0.4W-0.1-15ppm-Precision-metal-film-resistors/60260/kw/0.1%25#techspec

I agree with Ben that you can match to hundredths of ohms using 1%.  Buy a bag of 100 of each value and use a quality DMM to sort one value at a time into big square deli sized egg cartons.  You won't go through 50 before you have 8 matched to your tolerance.  Depending on the price of .1% it should be worth your time.  Make a jig to test resistance quickly without touching the leads.  I have to make one myself, and I am going to use two 2 or 3" 3/8 spiral springs with loops at each ends.  They are in any hardware store.  Screw one end of each to the thin side of a 2x4, and the other ends to a dowel handle.  The 2x4 ends have leads to the DMM.  I can flex the springs with the handle to insert and pinch my R for measurement.
Mike
PS: and for what it's worth, the most common low signal resistors I replace when fixing equipment are ultra precision resistors.  They drift in value, I think because they are more like "handmade".

I think he's building bridged T attenuators for stereo usage, with a Shallco 46 position switch for +/- 11dB in 0.5 dB steps.  So, yeah, count the number of resistors required, then think about precision matching. 

dig this
http://www.banzaimusic.com/MBB-SMA-Precision/

O.K I'm old and my memory sometimes fails me. But,  ;D
This is the company I have used on a couple of occasions. They are really friendly.



CTL Components Group plc
Falcon House
19 Deer Park Rd
London
SW19 3UX
United Kingdom

Switchboard +44 (0) 20 8545 8700
e-mail: [email protected]
web: www.ctl-components.com


Frank

So I made a jig and blew through my matching job:


I went with 1" springs because a 90 degree bend was easier.
Mike

Very clever, quick nice and easy check! Thanks for the Tip!

0.1% accuracy on small values with those springs plus the joint resistances between the spring + resistor and spring + leads would present a problem I would assume. I suppose you can measure the total loss and negate it against the final reading. Nevertheless it is a very smart jig.

Gold said:

This is for line amps and I want the absolute value to be the same for all eight attenuators.

Click to expand...

How about using a "linear" resistor chain and op-amp? As a bonus, errors in the absolute resistance are effectively cancelled out.



I know, it is not passive...

I already dropped the 1k USD on the IRC resistors. They are off to Shallco.

I want to use the line receiver-attenuator- makeup gain line driver topology because it allows for endless experimentation. I plan on doing a lot of listening tests with different amplifiers.

There are certainly less expensive ways to do it with the same quality and accuracy. They require skills that I don't have like laying out a circuit board.

Ouch, well quality costs