Mid Side M-S Matrix Uses No Precision Resistors

[emrr]

yep; you stepped in it!

[mediatechnology]

Just wanted to mention that MS boards and IC kits will be available again. I expect delivery around the beginning of the second week of April. viewtopic.php?f=6&t=346

[Gold]

On the subject of precision resistors; the only 0.1% resistors I can find in the catalogs in all the E77? values are the IRC ones in the Mouser catalog. I'm sure these are fine but are there any alternatives? I'm building a bunch of resistor string attenuators.

[Gold]

Thanks Roger. That's good to know. I don't have any preference for resistor brand. I guess if Holco was still available I'd go for those since they were widely used for audio. There are the Welwyn ones that are available through Newark but I didn't see all the 1% values in .1% I'm certainly not going to conduct listening tests. I hope they don't revoke my mastering engineers license.

[mediatechnology]

I would probably buy larger quantites of less expensive 1% MFs and sort to better than .1%.
Resistor quality has gone stratospheric in the last decade. Once you have compositioon you want (metal film) and exact value, you are there.

+1 on what rf said. I've seen very close matching myself with 1% MF through-hole parts. Buy 100 Xicors from Mouser and sort.

If you feel like SMT soldering onto a surfboard your R network you can easily get SMT 0.1% values from stock. Be careful which SMT you buy though - some suffer high THD 3 from self-heating. 0603 parts are particularly bad and on 0805's you need to pick vendors carefully.

[Gold]

Paul
I would probably buy larger quantites of less expensive 1% MFs and sort to better than .1%.

I have to build eight, 2 pole 45 position switches and they all should match in absolute value. I ain't sorting that many resistors. To me it's worth the money to just buy what I need.

[Gold]

Roger,

By absolute value I meant that I wanted all 16 decks to match. I could get a batch of resistors that could be matched to .1% but not .1% of the nominal 1% value. Another batch could match to .1% at a different value. Since it's a long resistor string I'd like predictable error tolerance. If any need to be replaced it would be a crap shoot. I know it's highly unlikely a resistor will fail but I feel better knowing I could just order the part and it would be right.

For the EE's I'm matching resistors. That's a manageable amount.

[mediatechnology]

I have to build eight, 2 pole 45 position switches and they all should match in absolute value.

Paul, that's serious. Given that I'd go for the 0.1% parts.

[Gold]

Paul, that's serious.

Yep. Monster Shallco's. Neither of the compressors for the console has make up gain front panel controls. All gain adjustments will be via the input and output amps. +/- 11db in .5dB steps. That should do it.

[emrr]

Hey Paul, sounds like a tough soldering job. I'm thinking that the actual ratio of values in what you're building is such that absolute precision isn't terribly necessary to get results that are indistinguishable with slight variance. Of course, you don't want to build and find out the hard way. I just see that many old Daven attenuators with much looser tolerances can be used interchangeably, and the standard 600/600 T decade attenuator always had a 0.1 dB selection, even way back when. What type of attenuators, and what values are we talking here?

I have some 1/3 dB 600 ohm H's that do 10 dB total.