Stereo to mono jack mixing circuit - what values of R1 to R3

Thread Starter

Mellisa_K

Joined Apr 2, 2017
391
Hello AAC people

Stereo to mono jack mixing circuit

1 Figure 1 shows my quesion: It includes a wiring diagram that is common on the internet. I would like your opinion please on what the value of resistors R1 R2 and R3 should be. Or at least clues on how I can work them out myself?



2 I want to feed my smart phone earphone output to my guitar amp mono input. The guitar amp is specified here: http://www.ashtonmusic.com/products/amplification/item/ga10 (except my amp being an apparently older model does not have the “ipod ready” input.

3 I want to achieve this by wiring the 3.5mm input male jack to another 12.5mm output mono male jack with the summing/ mising box in between.

4 The value of the resistors is the big unknown for me. Some specifications for this on the internet seem to be based on the assumption the circuit is to be used to feed output from a stereo souce into head phones. I found an AAC forum question in which @#12 and @AnalogKid @KJ6EAD specified resistor values based on this function https://forum.allaboutcircuits.com/threads/stereo-headphone-jack-how-can-i-make-it-mono.115611/ see posts for example #24, #11 #25

5 I assume that part of the goal of having these resistors in the circuit is to protect the input circuitry (from shorting the stereo chanels) not just to protect the destination amp from being overpowered by the doubled voltage of the input signal. If so couldnt you put a diode in place of R3? Or maybe since we are dealing with AC(?) the diode wouldnt pass the other half of the signal. If so would this problem be insoluable, ncessiating the use of the resisotor(s)? Either way i dont want to ruin the audio amplifier in my smart phone!!

6 Does the summing box need to be shielded?


Thanks

Mellisastereo to mono summing box.png
 

AlbertHall

Joined Jun 4, 2014
12,345
The things affecting the choice of resistors are: source impedance; load impedance; phone output voltage; and amplifier input sensitivity. We could make some reasonable guesstimates for the phone output but do you know the input characteristics of the amplifier?
 

Thread Starter

Mellisa_K

Joined Apr 2, 2017
391
The things affecting the choice of resistors are: source impedance; load impedance; phone output voltage; and amplifier input sensitivity. We could make some reasonable guesstimates for the phone output but do you know the input characteristics of the amplifier?
No I've only got the specs in the link I posted.

Two possible get arounds:
1 cld I physically measure it using my DMM?
2 Ashton is an Australian brand and I cld try to contact them by phone. I'd prefer I didn't have to cos companies these days won't let u near their technical staff or wait for their written response

Thoughts?
 

dendad

Joined Feb 20, 2016
4,452
I think you can go with what is in the circuit. In reality, the values don't matter that much as you will probably not be pushed for signal levels.
The 2 x 475R resistors will be harder to find so use 390R, or 470R, or for that matter, Even 1K. They are just stopping the output amplifiers being shorted together and allowing the signals to be mixed together without the other channel shunting the signal.You could go to the trouble of measuring things and calculating the correct values but what you have there will work ok.
 

Thread Starter

Mellisa_K

Joined Apr 2, 2017
391
I think you can go with what is in the circuit. In reality, the values don't matter that much as you will probably not be pushed for signal levels.
The 2 x 475R resistors will be harder to find so use 390R, or 470R, or for that matter, Even 1K. They are just stopping the output amplifiers being shorted together and allowing the signals to be mixed together without the other channel shunting the signal.You could go to the trouble of measuring things and calculating the correct values but what you have there will work ok.
Wow thanks Dendad. I'll have some loud music on wed arvo = next time off work. I'll let u know how I go
 

AnalogKid

Joined Aug 1, 2013
10,990
Are you plugging into a mic input or instrument input?

The two series resistors should be 1% values to mix the two signals at the same levels. The shunt resistor can be 5%, 10%, 20%, whatever, because it affects both signals identically. If possible, contact the amp people and ask what the input impedance is. Ideally, you want the series resistors to be less than 10% of that value.

ak
 

Thread Starter

Mellisa_K

Joined Apr 2, 2017
391
Are you plugging into a mic input or instrument input?

The two series resistors should be 1% values to mix the two signals at the same levels. The shunt resistor can be 5%, 10%, 20%, whatever, because it affects both signals identically. If possible, contact the amp people and ask what the input impedance is. Ideally, you want the series resistors to be less than 10% of that value.

ak
Instrument. There's only one input. The amp is marketed as a guitar amp.

Thx for the numbers AK I appreciate it. But...When you quote percentages what exactly is the denominator yre referring to? Surely not the voltage of the signal.
 

MrChips

Joined Oct 2, 2009
30,720
We are splitting hairs here.

I would go with lower values. R1 and R2 are the same, Make R3 about twice the value of R1.
Example, R1 = R2 = 100Ω. R3 = 220Ω 5%

1% or 5% represents the tolerance of the resistance.
For example, 100 1% could be 99-101.
or 100 5% could be 95-105.

1% or 5% is not going to make a difference in this case.

In reality, just about anything from 47 to 1000 will work. Just make R1 = R2. If you want, even R1 = R2 = R3.
 

Thread Starter

Mellisa_K

Joined Apr 2, 2017
391
We are splitting hairs here.

I would go with lower values. R1 and R2 are the same, Make R3 about twice the value of R1.
Example, R1 = R2 = 100Ω. R3 = 220Ω 5%

1% or 5% represents the tolerance of the resistance.
For example, 100 1% could be 99-101.
or 100 5% could be 95-105.

1% or 5% is not going to make a difference in this case.

In reality, just about anything from 47 to 1000 will work. Just make R1 = R2. If you want, even R1 = R2 = R3.
Oh ok thx that's clear now. Thx mrchips
 

Thread Starter

Mellisa_K

Joined Apr 2, 2017
391
Are you plugging into a mic input or instrument input?

The two series resistors should be 1% values to mix the two signals at the same levels. The shunt resistor can be 5%, 10%, 20%, whatever, because it affects both signals identically. If possible, contact the amp people and ask what the input impedance is. Ideally, you want the series resistors to be less than 10% of that value.

ak
Yes I got it. You want the two resistors to have minimum variance and that's important because the sound levels must be as close to each other as possible. And statistically gold band says in their manufacture there's least chance their actual value will differ from their coded value. I got there. Yes! Doh. Mrchips got us there AK!
 

AnalogKid

Joined Aug 1, 2013
10,990
Actually, what I'm saying is that, depending on your hearing, gold band (5%) tolerance is not good enough for the series resistors.

Also, there are adapter cables you can buy on ebay with the resistors built in.

ak
 

Thread Starter

Mellisa_K

Joined Apr 2, 2017
391
Actually, what I'm saying is that, depending on your hearing, gold band (5%) tolerance is not good enough for the series resistors.

Also, there are adapter cables you can buy on ebay with the resistors built in.

ak
Yes I got that I was just quoting what I thought was the lowest tolerance, gold; but as you say that's 5% so what colour is 1%? I thought there was only silver (with 10%) apart from gold.
 

#12

Joined Nov 30, 2010
18,224
I say the resistor tolerance is not very important. First, human ears can't be sure there is much of a difference until a 3db difference exists, and that is about 30% or 40% of the voltage, depending on which way you hold your calculator. Second, if the two input values aren't the same, use the balance adjustment.

With different information on each channel (the definition of stereo) and the vagueness of the human ear, I don't think it's possible to twist a knob to 1% just by listening.
 

Thread Starter

Mellisa_K

Joined Apr 2, 2017
391


Go with gold. It will be OK.
Go with silver and you still would not hear the difference.
Oh so there are 5 more stringent tolerance bands: br, red, green,blu,purple. The only charts I've seen have so far shown only the two - gold and Silver
 

Alec_t

Joined Sep 17, 2013
14,280
I agree with #12 and don't think the resistor tolerance matters greatly. I doubt that the phone output channels or the guitar amp channel gains are well matched.
 

AnalogKid

Joined Aug 1, 2013
10,990
human ears can't be sure there is much of a difference until a 3db difference exists,
Actually, Bell described one decibel (one-tenth of a Bel) as the smallest change in volume a human can detect (steady state tone, no hearing defects). Bell was an audiologist. His initial interest in electricity was not to revolutionize world-wide communications; he was working on a hearing aid.

ak
 

MrChips

Joined Oct 2, 2009
30,720
Actually, Bell described one decibel (one-tenth of a Bel) as the smallest change in volume a human can detect (steady state tone, no hearing defects). Bell was an audiologist. His initial interest in electricity was not to revolutionize world-wide communications; he was working on a hearing aid.

ak
That may be correct.
But look at this application. We are mixing (summing) LEFT and RIGHT channels into MONO.
LEFT and RIGHT channels most likely have 50% of the same audio content. Differences are for stereo effect and enhanced musical enjoyment.
If you lose 20% of either channel who is going to notice what you are missing.
 

#12

Joined Nov 30, 2010
18,224
Bell described the decibel (one-tenth of a Bel) as the smallest change in volume a human can detect (steady state tone, no hearing defects).
Right. A trained listener with steady tones in a laboratory environment clocked humans at 1.1% of the voltage ratio.
Now, try that in front of a guitar amplifier with real music containing a peak to average ratio of ten to one, with two channels containing different information. Have you ever thought to yourself, "I should tell Paul McCartney to turn his volume knob up by 1%?"

The only charts I've seen have so far shown only the two - gold and Silver
We're trying not to dump it all on you at once.:D (You probably won't be doing anything so special that temperature tracking will be important in the first year.)

Now for some theory. Electric guitars often contain volume and tone controls in the range of 250,000 ohms, so guitar amplifiers are high impedance inputs, like 680,000 ohms. If your music supply has an impedance of less than 68,000 ohms, the guitar amplifier will believe that's just fine.

The other end is the music supply. Earphone outputs are usually designed to drive something way less than 100 ohms, so you can protect the music source with anything from 100 ohms to...68,000 ohms. Don't go to the 68,000 ohm end. It tends to pick up noise. I'm going with MrChips in the 100 ohm to 1000 ohm range.
 

Thread Starter

Mellisa_K

Joined Apr 2, 2017
391
Right. A trained listener with steady tones in a laboratory environment clocked humans at 1.1% of the voltage ratio.
Now, try that in front of a guitar amplifier with real music containing a peak to average ratio of ten to one, with two channels containing different information. Have you ever thought to yourself, "I should tell Paul McCartney to turn his volume knob up by 1%?"


We're trying not to dump it all on you at once.:D (You probably won't be doing anything so special that temperature tracking will be important in the first year.)

Now for some theory. Electric guitars often contain volume and tone controls in the range of 250,000 ohms, so guitar amplifiers are high impedance inputs, like 680,000 ohms. If your music supply has an impedance of less than 68,000 ohms, the guitar amplifier will believe that's just fine.

The other end is the music supply. Earphone outputs are usually designed to drive something way less than 100 ohms, so you can protect the music source with anything from 100 ohms to...68,000 ohms. Don't go to the 68,000 ohm end. It tends to pick up noise. I'm going with MrChips in the 100 ohm to 1000 ohm range.
i am sooo glad i have got you guys to refer to. whenever i ask the simplest question, i learn the answer AND i get lots of other information that i can file away. things begin to click. i like clicking, joining dots and making sense of disparate bits of information. we all do. its humanity. obsessive problem solvers and problem finders. good to be alive eh?
 
Top