Hi everyone!
I am working on a circuit that will allow me to possibly record iPhone conversations using a hardware device instead of apps that are unreliable or require a rather costly service. I need to conduct interviews and this is a long-needed circuit.
My main problem is with trying to figure out how to combine the L & R channels with the MIC and/or Audio Input without the L&R channels bleeding over into the MIC output (the Red Jack).
Any simple solutions?
I'm a complete amateur, so I could use all the help I can get. Thanks!
I am working on a circuit that will allow me to possibly record iPhone conversations using a hardware device instead of apps that are unreliable or require a rather costly service. I need to conduct interviews and this is a long-needed circuit.
My main problem is with trying to figure out how to combine the L & R channels with the MIC and/or Audio Input without the L&R channels bleeding over into the MIC output (the Red Jack).
Any simple solutions?
An externally hosted image should be here but it was not working when we last tested it.
I'm a complete amateur, so I could use all the help I can get. Thanks!
Last edited:
Ok, I think I've solved it. It seems it was as simple as adding R18, R19 and R20, a group of 10K resistors. Does this circuit look ok? I'm open to critique ...
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
C3 & C5 looks to be the wrong way around, and do you really need 20dB of gain in the recording buffer amplifier?
R13 does nothing meaningful (It is in series with the 10K summing resistors so it just reduces the isolation level).
Switching the power to the input amplifiers is a going to give almighty thumps if operated with the system live.
Both input amplifiers have 40dB of gain which is surely excessive for the line input and may be excssive for the condenser mic element (Given you are driving a mic input anyway), U2 is probably redundant and U1 may be redundant depending on the other details.
Just what I see in a quick look.
Regards, Dan.
R13 does nothing meaningful (It is in series with the 10K summing resistors so it just reduces the isolation level).
Switching the power to the input amplifiers is a going to give almighty thumps if operated with the system live.
Both input amplifiers have 40dB of gain which is surely excessive for the line input and may be excssive for the condenser mic element (Given you are driving a mic input anyway), U2 is probably redundant and U1 may be redundant depending on the other details.
Just what I see in a quick look.
Regards, Dan.
I'm honestly not sure. From my limited understanding, they are just to neutralize the input voltage. I could probably do without C5, right?
Not sure, I'll let you know once I finish building it and give it the "trial and error" test. I mainly want to be able to control the volume level at each point so the mixed output is able to be picked up regardless what device it's connected to for recording.
Excellent point! Thanks for pointing this out.
Like previously stated, I don't want to build a circuit that's too weak and doesn't serve it's purpose accordingly. I'd rather it be "too much" and keep it turned down than "too little". Just my thinking based on my limited knowledge, although it may be wrong. The condenser mic DOES work great as wired in this diagram, that much I have tested thus far.
U2 was added to give the ability to "boost" the line-in if it's too low. It might be unnecessary, but it gives greater control over the level and there for that possible occasion the input is too weak.
THANK YOU!
I really appreciate you taking the time to look at this and giving your feedback. Like I said, I don't know much about the circuit and it's kind of a combination of other circuits combined along with what little understanding I have of it. Since there isn't a circuit that does this, I'd like to make it as best as possible for the benefit of others, so I'm open to suggestions/improvements.
C1 - C5 serve to isolate the DC operating point of the opamps (set by the resistor networks on the non inverting inputs) from the audio IO which is reference to ground, you need C5 it is just the wrong way around.
The opamps have a reference set at the supply volage times the potential divider ratio fromed by the resistors connected to the non inverting input, and the caps should be oriented to be correctly biased given this fact. I would guess the dc operating point is at half the supply rail but you dont specify the resistor values here.
The trouble with providing 40dB of gain before the volume control is that it will be very easy to clip the input amplifier, and that turning down the volume will not help. If for example vcc is +12V, and the opamp can swing maybe 5V either side of its mid rail supply, then with 40dB of gain you are looking at only 50mV to clip the line input....
Make R8 10K - 100K and the thing will be better behaved.
741's? Really, noisy, and have not been seen in the wild for the last 20 years, try something like a '071 '072 or 074 (single, dual and quad respectively) especially if this thing is going to running off a battery.
R7 could usefully be 470 ohms.
The opamps have a reference set at the supply volage times the potential divider ratio fromed by the resistors connected to the non inverting input, and the caps should be oriented to be correctly biased given this fact. I would guess the dc operating point is at half the supply rail but you dont specify the resistor values here.
The trouble with providing 40dB of gain before the volume control is that it will be very easy to clip the input amplifier, and that turning down the volume will not help. If for example vcc is +12V, and the opamp can swing maybe 5V either side of its mid rail supply, then with 40dB of gain you are looking at only 50mV to clip the line input....
Make R8 10K - 100K and the thing will be better behaved.
741's? Really, noisy, and have not been seen in the wild for the last 20 years, try something like a '071 '072 or 074 (single, dual and quad respectively) especially if this thing is going to running off a battery.
R7 could usefully be 470 ohms.
C1, C3 and C5 are all in backwards.40 dB of gain on the mic might be a little low but it's in the ballpark. 40 dB on the line input is way too much. I'd run that around 6 dB or less and you still run the risk of driving U2 into clipping if the power supply is only 9V. The record out amp should have minimal gain. Unity to +3dB. There should also be small value (22 to 100 ohm) resistors in series with C2, C4 and C6 to isolate from capacitive loads with C6 being the most important.
The line level input impedance is 1K and is too low. Change R2 to 10K and R3 to 15K to drop the gain to 3.5 dB and raise input impedance to 10K. Change R16 to 15K also to reduce the ouput buffer gain and you may find it needs to be 10K or even less.
Agreed on 741 amps. They prove it works but better anps would be a big improvement.
The 'bleeding' is very likely the amps driven severely into clipping.
G²
Thank you all very much!
I'm a bit baffled by how you all are calculating the gain ... according to this gain calculator, 741 Op Amp Gain Calculator, it is figured by dividing Rf by Rin. In my diagram, it would be R3/R2 = -100. Not sure where the 40dB comes from.
I've got some TL071 Op Amps ordered and will try to rebuild the circuit using them instead since you all have unanimously said they are a better choice.
What I am wondering is if any of you can possibly point me in the right direction on what I need to learn and know to build/design the new circuit correctly. I wish even more so now that I had been able to finish tech college on electronics engineering years ago, but simply couldn't afford it. I don't mind learning though, but lack knowing where to start. I was an electrician for about 4 years and excelled in school for it, but it gave little help to understanding this sort of stuff.
So, thanks again for your help and any information regarding on redesigning it with the TL071 chips instead. If any of you are feeling generous enough to help and teach me to better understand the stages of this project, I would be forever grateful for it. 😉
I'm a bit baffled by how you all are calculating the gain ... according to this gain calculator, 741 Op Amp Gain Calculator, it is figured by dividing Rf by Rin. In my diagram, it would be R3/R2 = -100. Not sure where the 40dB comes from.
I've got some TL071 Op Amps ordered and will try to rebuild the circuit using them instead since you all have unanimously said they are a better choice.
What I am wondering is if any of you can possibly point me in the right direction on what I need to learn and know to build/design the new circuit correctly. I wish even more so now that I had been able to finish tech college on electronics engineering years ago, but simply couldn't afford it. I don't mind learning though, but lack knowing where to start. I was an electrician for about 4 years and excelled in school for it, but it gave little help to understanding this sort of stuff.
So, thanks again for your help and any information regarding on redesigning it with the TL071 chips instead. If any of you are feeling generous enough to help and teach me to better understand the stages of this project, I would be forever grateful for it. 😉
It's 6th grade arithmetic if you remember logarithms. Gain in dB is 20 log absolute value of gain. Log of 100 is 2 times 20 is 40dB. That wasn't hard.
Unless you're dealing with low level signals from mics, phono cartridges or optical sensors, it's rare to need gains over 10 ( 20 dB ). Remember too that gain need not be voltage. It can also be current gain to drive heavier current loads like speakers or motors - another way of saying you reduced the output impedance of the amp.
If you read 'The Art of Electronics' by Horowitz and Hill, you'll know more than a lot (most?) of the people here. Also a lot of good info in the ARRL handbook. The white papers at Analog Devices are free and very informative. Ron Mancini's 'Opamps for Everyone' is also free on the TI site.
G²
Ok, so I've come to realize my original post was approaching the issue in way too complex of a manner and I have not resolved the basic issue at hand. So here's a diagram of what I'm trying to accomplish WITHOUT any op-amps, etc.
In the diagram, the headphones should ONLY hear what is coming in from the microphone and not the MP3 player. On the headphone plug, both audio sources should be combined. The problem is I'm hearing the mp3 player on the headphones since the 10K ohm resistors do not block the MP3 sound from bleeding back onto the microphone channel. Any ideas how to do this correctly?
EDIT: They are all electrically connected via ground if that has any implications. The lines shown here would be for the positive lines.
In the diagram, the headphones should ONLY hear what is coming in from the microphone and not the MP3 player. On the headphone plug, both audio sources should be combined. The problem is I'm hearing the mp3 player on the headphones since the 10K ohm resistors do not block the MP3 sound from bleeding back onto the microphone channel. Any ideas how to do this correctly?
An externally hosted image should be here but it was not working when we last tested it.
EDIT: They are all electrically connected via ground if that has any implications. The lines shown here would be for the positive lines.
Last edited:
A virtual earth summing stage, using an op amp, will do this. Why not use an op amp? There are some fantastically clean amps available now for a pittance. I say use them! Yes, you need a supply and a layout that does not suck, but the end result will work quite well.
@Monte, the reason I was saying "without an op-amp" is to do it in the simplest way possible for better understanding, then use op-amps for better performance, etc.
As far as using an op-amp for this, would using the TL071 and connecting the two different inputs to pins 2 & 3 be the answer? (I need to read up on the difference between inverting and non-inverting inputs...)
So say I connect the mic input to pin 2 and the MP3 player to pin 3 ... is this correct? Thanks for the help!
As far as using an op-amp for this, would using the TL071 and connecting the two different inputs to pins 2 & 3 be the answer? (I need to read up on the difference between inverting and non-inverting inputs...)
So say I connect the mic input to pin 2 and the MP3 player to pin 3 ... is this correct? Thanks for the help!
I'm pretty sure that's incorrect. But here's what I don't understand ... the summing op amp circuits I've found combine the two channels like the way I did in the illustration, which will have bleed over into the microphone channel. Here's an example circuit:
So V1 and V2 are connected and if I were to connect my headphones to V2 before R2, I'd hear the audio from V1 and that's what I'm trying to solve ... how to separate them and yet combine them later on. I'm a visual learner, so that's why I use illustrations. It helps me understand it.
So V1 and V2 are connected and if I were to connect my headphones to V2 before R2, I'd hear the audio from V1 and that's what I'm trying to solve ... how to separate them and yet combine them later on. I'm a visual learner, so that's why I use illustrations. It helps me understand it.
The MP3 player level is likely 40 to 50 dB higher in level than the mic. A passive mixer won't cut it here. At the least you'll need a mic preamp to get the level roughly the same as the MP3 player and _then_ you can finish with the summing amp suggested by Amereservant. That amp can drive the headphones.
G²
G²
You could be a jerk and use the one op amp inverting summer as shown above with a 1K ohm resistance for the mike input, 100K ohm for the feedback resistor, and 50K ohm input resistor for the MP3 player. That way, the same stage gives you 40dB gain for the mike with a pretty close to standard loading, a gain of 6dB for the MP3 and whatever drive you'd need according to the opamp's capabilities.
Sorry that I haven't studied the details to know the exact routing you'd need re. stereo etc., but the idea here is that you can use the inverting stage as a multi-purpose stage, without requiring a separate mike amp circuit. The penalty is extra noise and distortion on the MP3 input, but that should not be an issue with a good op amp. BTW, an NE5534 would be a better choice for this sort of stage, compared to a TL071. The 5534 is pretty well behaved when it's run as an inverter, and practically, the noise specs will be not all that shoddy with the 1K input mike amp resistor.
Sorry that I haven't studied the details to know the exact routing you'd need re. stereo etc., but the idea here is that you can use the inverting stage as a multi-purpose stage, without requiring a separate mike amp circuit. The penalty is extra noise and distortion on the MP3 input, but that should not be an issue with a good op amp. BTW, an NE5534 would be a better choice for this sort of stage, compared to a TL071. The 5534 is pretty well behaved when it's run as an inverter, and practically, the noise specs will be not all that shoddy with the 1K input mike amp resistor.
- Status
- Not open for further replies.