Resistor noise....

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

alk509

Well-known member
Joined
Sep 3, 2004
Messages
1,207
Location
MA, US
Hey kids,

We're struggling to decide which resistors to buy for our 1G resistor group buy... We have the option of buying metal oxide or metal glaze resistors. The metal oxide ones are cheaper than the metal glaze ones, but what about noise???

So far I haven't found anything as far as how the noise specs of these two kinds of resistors compare. I think metal glaze is just some Bri'ish name for metal film, but I really know nothing...

So what's noisier, metal ox or metal glaze? And does it even matter for our application (condenser mics)?

Thanks!

Peace,
Al.
 
I don't think we need to consider noise here. After all, the 1G is shunted by capsule capacitance. I've tried some different types, and had NO difference in noise figure. Noise difference when trying different EF86's is much, much larger.

Jakob E.
 
Jakob, is shunt capacitance an issue with these resistors? What is the nominal capacitance of a condensor capsule, and would a few pF of shunt capacitance in the resistor make a difference?

I'm curious, because you can reduce shunt capacitance by putting a few lower valued resistors in series (A trick I learned from Bob Pease's Troubleshooting Analog Circuits).

Cheers,

Kris
 
Kris,

The capacitance of the capsule is somewhere from 70 to 140pF.

Any shunt capacitance to ground will act as a capacitative voltage divider, not as a filter as usually.

This is how front-end pads on condenser mics are most often made - shorting signal out to ground via a small capacitor...

Jakob E.
 
Hi, new to the forum.. not entirely sure where to post this, so I figured I'd tack it onto the end of a resistor thread... but, is there a reason the bulk of people are using large axial components and avoiding surface mount? Have people had performance problems, or is it just for the simplicity of assembly? (which is kind of subjective).
I ask because it is certainly very easy to pack all of the components for such a mic design into a very small area using surface mount parts.

Cheers,
Sean
 
your sig is peculiarly relevant to that thread :)

so it seems the consensus on resistors is BIGGER IS BETTER... But if board space is at a premium, thin film is better than thick film, and metal foil sounds like an even better option.

Cheers,
Sean
 
Metal foil is about as good as it gets from a linearity/low noise/low tempco standpoint. But it is difficult to make very high value parts, so we have to take what we can get there. As mentioned before there are some high value SMD resistors available but they can be very very noisy.
 
they after blow up the resistor? or they buy a 7G resistor and paint the line :oops: sorry for the bad joke.

I know the value is not critical but how we can measure that resistance?
 
That interesting I was going to make a pcb with two solder pads and test different pencil leads to see what I could make. Need to read read up on china ink. Thanks for that info.
 
Do you need current to make noise?
How much current flows thru the 1 gig resistor in a typical mic circuit?
What would be the max voltage across that resistor?
 
> But if board space is at a premium

Then your priorities are wrong.

"Sound" and "Maintainability" are more important than board space.

> they after blow up the resistor? or they buy a 7G resistor and paint the line.

Call your regular supplier of crates of resistors, ask them to build some customs that don't get any carbon/metal applied, just a ceramic body with two leads.

Then put a tiny bit of carbon on. Pencil, aquadag (carbon, alcohol, and a binder), or what we here used to call India Ink (plain drafts-man's or fountain-pen ink) which is mostly water and carbon, with a little ox-gall binder.

> how we can measure that resistance?

Not easily; but special tubes or ordinary FETs can read nanoAmps. 70V across 7Gig gives 10nA. Or another way: a 7Gig and a 7Meg makes a 11:1 divider, put 100V across it and see if you get 9V out. With common bench VTVMs, you have to deduct the meter's 11Meg input resistance from the 7Meg reference, just math.

Or put it in the head-amp with a dummy capsule. If the cathode voltage is wrong, the resistor has too many Gigs; if the bass response slumps or the bass-noise is high, the resistor is too small. You don't need to know the resistor's actual value, just that it gives your best compromise between draining grid leakage and injecting thermal noise into the system.

> How much current flows thru the 1 gig resistor in a typical mic circuit?

Grid leakage. Less than a nanoAmp.

> What would be the max voltage across that resistor?

Same as the basic body that you drew it on. 1/2W resistors are usually rated 250V or 400V. Some of that is the carbon, but go much over these numbers and the ceramic may break down. This is utterly unimportant in mike use, unless you are testing the resistor with brute force. (7,000V across 7Gig would give an easy to measure 1mA, but probably an arc.)

> Do you need current to make noise?

Atoms won't stop wigging, or not at any temperature you want to work in. So any resistance has a noise.
 

Latest posts

Back
Top