[Duran]

I have some questions for building an attenuator, specifically, on the impedance the connected devices will see from it.
 
 
I want to connect a Line level source to a mic input. So I want to make an attenuator like one of these. Because I didn't see any mention of advantages from one design to the next (besides differences in what value resistors may be conveniently available), I thought I might as well just do the basic L design. And since I can't imagine how hard it'd be to find a dual-gang pot that would properly adjust both resistors through the proper ranges together, I'll have to go with fixed attenuation. (which is a shame, because I do have tons of potentiometers. Many are probably designed for this already, but I don't know how to find one that would have the correct resistance/impedance values/ranges and drop in dBs)
 
 
The device I want to put the source into (the mic input) actually originally had a microphone built into it. It was rated at 2K ohm impedance, so I assume the new source (+attenuator) should have the same, right? I assume that's what the device is "expecting". So I'll need to know the impedance of the source signal so I can determine what values the resistors should have to come to that 2K goal (while simultaneously reducing the decibel levels of course), right? How do I find that? And should the source be active during measurement? Or is this with the circuits off? What happens if I measure it the opposite way of that?
 
So basically I need help determining the values for these resistors (and perhaps how to find pots among the spare parts I have that will work, if that's not as unlikely as I may think). Apparently the decibel drop for Line to Mic levels is ~40-50. That's the only variable I know besides the impedance of the original mic.
 
Any help would be very much appreciated.

 

[Duran]

Bump.

 

[Duran]

Hmm... is there perhaps a better place to post this?

 

[xxbaker]

Quote:

I have some questions for building an attenuator, specifically, on the impedance the connected devices will see from it.
 
 
I want to connect a Line level source to a mic input. So I want to make an attenuator like one of these. Because I didn't see any mention of advantages from one design to the next (besides differences in what value resistors may be conveniently available), I thought I might as well just do the basic L design. And since I can't imagine how hard it'd be to find a dual-gang pot that would properly adjust both resistors through the proper ranges together, I'll have to go with fixed attenuation. (which is a shame, because I do have tons of potentiometers. Many are probably designed for this already, but I don't know how to find one that would have the correct resistance/impedance values/ranges and drop in dBs)
 
 
The device I want to put the source into (the mic input) actually originally had a microphone built into it. It was rated at 2K ohm impedance, so I assume the new source (+attenuator) should have the same, right? I assume that's what the device is "expecting". So I'll need to know the impedance of the source signal so I can determine what values the resistors should have to come to that 2K goal (while simultaneously reducing the decibel levels of course), right? How do I find that? And should the source be active during measurement? Or is this with the circuits off? What happens if I measure it the opposite way of that?
 
So basically I need help determining the values for these resistors (and perhaps how to find pots among the spare parts I have that will work, if that's not as unlikely as I may think). Apparently the decibel drop for Line to Mic levels is ~40-50. That's the only variable I know besides the impedance of the original mic.
 
Any help would be very much appreciated.

A potentiometer is just a set of variable resistors all controlled by a single wiper.  It controls a variable resistor between source and output and one between source and ground (and then mirrors those two on the other channel).  So wiring up a potentiometer would be quite easy in this scenario, the problem is getting 50db of drop is most of the range of the pot so you're probably better off just using regular resistors and adjusting volume of the source to account for that.
 
Also I'm not sure you really understand the significance of impedance (I'm not sure I really do either =P).  I'm assuming what it means by "rated at 2K ohm impedance" means the input impedance of the mic in is 2K ohm, not anything you input into the mic in should be 2K ohm and even if it was, it would likely be anything you input should be less than 2K ohm.  When connecting components if you keep a ratio of 1:10 output:input ratio you're fine.  So if the mic in has an input impedance of 2K ohms you'll want the output impedance of your device to be 200 ohms or less.
 
The simplest thing to do would be to calculate out the resistors for a 50db drop and wire that up and use it.  If you're really worried about matching impedances, run a sweep on the source and record it and check frequency response.  That way you'll be assured it's all working properly.

 

[Duran]

Yeah I know a potentiometer would be easy to wire, but finding one that is logarithmic and has the right ranges on BOTH sides of the wiper seems like quite a task. Although most likely because I don't even know the values or ranges I need to find. But also because I'm pretty sure that the ranges on either side need to be different. I could put resistors on a side, but I don't think the two ranges even vary by the same amount, so I don't think it'd be enough for one to simply be at a lower resistance range.
 
 
So you mean you think the circuitry of the device the mic plugs into is what has the 2K ohm impedance? I thought that was the impedance the mic itself presented to that circuitry, it's resistance + inductance. I think that output of the attenuator (input connected to audio source) should have the same impedance as the mic would, right? Which I believe would be 2K ohms.
 
 
Also, this audio source I want to attenuate and plug into this other device's mic input is 'supplying it's own power', if you know what I mean. A microphone such as the one the device was designed with is just a passive component, powered by the device's own circuitry. I don't really know if that makes a practical difference or if it matters. I assume it doesn't, but it raises questions in me (e.g. is the device (or part of it's circuitry) now the load in the circuit as a whole since it's connected to the output of this audio source? (whereas before, the mic would be the load)). Maybe that's why we were looking at "which is the 2K ohm device?" differently.
 
You say the simple thing to do would be calculate the values for the resistors, but that's really the crux of my problem; I don't know how to find those. There are examples on that page I originally linked to, but I couldn't follow Example 1. For one thing I don't know what "K factors" are... it seemed to assume I already knew.

 

[xxbaker]

Well since a pot is designed to do exactly this (pad an input by varying amounts) I don't see why you're worrying about getting the right values so much.  Pots are designed based off of the equations used to calculate the fixed resistors in the link you gave, so I don't see what the problem is.
 
I don't really know about the impedance thing, but I assumed it was the input impedance of the device for these reasons:
1. The device is constant, but mics can vary, so when they give a single value like it's a fact it implies it's for the device
2. If it's not for the device and it is meant for the microphone, why would they say 2k as opposed to less than 2k.  If you have bridging impedance matching when a mic is at 2k, then a mic with any less (most mics are much less than 2k) would also work - so why don't they state that?
 
Of course I could be completely wrong about that. As far as phantom power goes, you'll want to turn that off when you're connecting your source to the mic input.  Also mics aren't really being "powered" by what they're being plugged into (other than phantom power).  The mic input looks for voltage differences on the input and then records based on those differences - or more accurately records whatever voltage differences there are on the input, if there's none then it records as no signal.  A mic converts physical energy to electrical energy passively ('active' microphones have an amp of sorts built into them to boost the signal when the mic would normally output much too low of a signal).  This means that when you blow on a mic, there's a changing voltage difference on the output of the mic.  It's not really like the mic plugs into the mic in and then they work together, I mean yes that's how it is but they both work separately also, so you don't need to worry about what is the load because either way whatever is being driven (the mic in) is the load and whatever's supplying the signal (the mic or your source) is the source.
 
K-factors are just logarithmic calculations.  You don't need a table to do the calculations - it's just if you don't have a calculator then you could use a table instead. So to do a 50db drop with 2K output impedance, your k-factor is ~316 (get that by doing 10^(50/20) or 10^2.5 or ~316).  So if you want your output impedance to be 2K, then your Rshunt is 2K ohm, and your Rseries will be ~630K.  
 
50db drop with 100 ohm output impedance, your k-factor is ~316.  Rshunt is 100, Rseries is ~31.5K.  Does that all make sense?  If you wanted to do a U pad instead of an Lpad, your Rshunts would be the same, but your two Rseries resistors would be divided by two (~315k and ~15.75k respectively).

 

[nikongod]

It sounds like you are looking for another input on a device that has all of its inputs used.
 
Why not get an external source selector for 4 sources into 1 (for example, other switches are available, but radio-shack sells a 4 into 1 source selector for about $25) and use that? It is much simpler and WILL work out better in the end. I have never tried to build a "cost is the only object" source selector (heh, mine is unplugging the cable, for free!)but I dont think it can be done for $25 unless you get creative with cardboard and takeout food containers. Even the $25 source selector is better than attenuating signal by XXdb for the express purpose of amplifying by XXdb.
 
On that note, by attenuating the output of your source to use the microphone input you are attenuating specfically so you can use more gain. At the minimum the noise pickup will be compromised (more noise pickup) , and you will pick up distortion from the added active components too. The only reason this may be a good idea is if you are using a CD player to test a mic input as you describe with better control than playing a speaker into a microphone or a phono input with a "de-riaa" board with fewer variables than the full turntable setup.

 

[Duran]

Ahh, ok. Those calculations make more sense now. I'll try the resistances for the 50 dB drop with 2K ohm impedance.
 
I still don't know about the output impedance question. But some systems may not be designed for use with different mics.
 
I'm not familiar with the term Phantom Power; I've heard it before but don't know what it is. I've never taken an electronics class, so I've just been teaching myself so far. Why is it important my source is turned off before connecting it?
 
And yeah, I know mics produce voltage as they react to sound... I guess I just assumed it was accompanied by additional power from the device for some reason. So, even though natively all the input (mic) is is a coil of wire and nothing else, the entire rest of the circuitry (the device), is still all considered the load? That seems odd, since that's where the amps, power source, and everything else is. Although if there's pretty much no power besides the mic generated voltage/signal in the immediate part of the circuit concerning the mic (before the first stage of amplification), I suppose that makes more sense since you could say "nothing's happening" before the mic.
 
As for the pots, well I was thinking that the resistor values would each need to vary by different amounts, and that a potentiometer's two value ranges (on either side of the wiper) would be the same (since there's usually one common track/fixed contact). Also, I'm not sure how I'd find an appropriate one, partly because I don't know if one that can go too low in attenuation would damage anything. So really because I just don't know what resistance ranges I would be looking for either side of the pot.
 
Thank you very much for your help btw. And sorry it took me so long to respond... I didn't forget, I've just been fairly busy lately. ^_^`