You are in the right forum. You just don't realise yet that audio electronics is not quite as trivial as you imagined. To get from where you are now to being able to design what you need will take you several years of study and experience. If you can invest that time then you will be rewarded with an interesting hobby. If you can't, then you need to look for a kit or even a ready-made item.
You have got a lot of good advice.
My 0.02-bitcoin---
After decades of DIY audio, going back in the 1960 dawn of discotheque (not Disco!) there are some things NOT worth doing yourself.
A GOOD cross-fade is one of them. There are many subtleties. What do you want the mid-point to do? Full-up on both? -3dB (constant power)? -6dB (constant peaks)? -4dB (a popular compromise)? or -20dB on both (fade to grey and back up the other side)?
A constantly-used cross-fade WILL wear out. I dunno why moreso than simple level pots, but mature (late 1980s) DJ crossfades used opto or VCA to get acceptable use-life. Maybe not an issue if you only cross-over every hour?
What you describe IS a "DJ mixer". Just ignore extra frills (voiceover etc). If you work with DJs, talk to them, some will have a fine basic mixer set-aside in favor of some newer sexier model.
My 0.02-bitcoin---
After decades of DIY audio, going back in the 1960 dawn of discotheque (not Disco!) there are some things NOT worth doing yourself.
A GOOD cross-fade is one of them. There are many subtleties. What do you want the mid-point to do? Full-up on both? -3dB (constant power)? -6dB (constant peaks)? -4dB (a popular compromise)? or -20dB on both (fade to grey and back up the other side)?
A constantly-used cross-fade WILL wear out. I dunno why moreso than simple level pots, but mature (late 1980s) DJ crossfades used opto or VCA to get acceptable use-life. Maybe not an issue if you only cross-over every hour?
What you describe IS a "DJ mixer". Just ignore extra frills (voiceover etc). If you work with DJs, talk to them, some will have a fine basic mixer set-aside in favor of some newer sexier model.
I was confused as to why you would want to blend from one source to another. Most home use is one source or another. I thought maybe you wanted to play your guitar along with your music.
Since you appear to have a DJ mixer, why do you want to add another mixer in line after it? The DJ mixer should be able to mix whatever your sources are and send them at a decent level through fairly long cables.
With proper choice of low current op amps they could last quite a while. However, low current opamps go against being able to drive long lines.
I'm sure that you can find a regulated power supply kit or pre-assembled that uses 3 pin regulators and just a couple parts. With practice, I can now assemble a discrete regulated +/-15V supply in about an hour. It's not that big a deal.
Your understanding of resistors is incorrect. That may stem from reading about current limiting resistors in an LED circuit. In operation, an LED has a fairly constant voltage across it, say 1.2V for an old fashioned red LED. To run that from a 15V supply you choose a resistor that will drop the 13.8 V at the desired current. Say you want 5 mA to get the desired brightness of the LED. V=I*R, solving for R=V/I. In this example you get 2760 Ohms. It's not that the resistor dumps excess voltage to the LED, it simply limits the current. If the voltage applied was higher or lower the LED would get brighter or dimmer, not switch on and off.
That's where the comparator comes in. It's a type of op amp circuit (there are opamps specifically designed for use as comparators but general purpose opamps will do for this application.) that compares an input voltage to a reference and if the input exceeds the input the output goes high or low. In effect, it is a voltage controlled switch. You need a reference to tell it when to turn on. Since you want to use this as a clipping indicator you want the LED off until the signal exceeds the set point. It wouldn't be particularly useful to have to decide if that's the brightness that indicates clipping or is it just below (making your eye the comparator). Once you wrap your mind around it, a dual op amp and half dozen parts is a pretty simple way of making this happen.
Cool. I think I'm just starting to get this stuff. I spent the whole weekend going over high school physics. For the first time in my life, I'm just starting to understand about current and voltage. Crazy huh? [emoji28]
Everything you just said actually made complete sense. If you'd told me that all last week, I wouldn't have understood what the hell you're on about. Haha.
Thanks Mr Bob Ellis.
Sent from my iPhone using Tapatalk
So now I have another few questions that you guys can help me with:
What are the nominal signal level and maximum signal levels (in terms of actual voltages) I can expect from a typical piece of pro audio gear such as a DJ mixer? And then how much gain should be applied to that by the first op amp in my mixer? I take it that the first thing that input signal is going to encounter will be the op amp.
Sent from my iPhone using Tapatalk
What are the nominal signal level and maximum signal levels (in terms of actual voltages) I can expect from a typical piece of pro audio gear such as a DJ mixer? And then how much gain should be applied to that by the first op amp in my mixer? I take it that the first thing that input signal is going to encounter will be the op amp.
Sent from my iPhone using Tapatalk
Pro gear often uses the reference 0dBu
0dBu = 0.7746Vac
+4dBu = 1.585 * 0.7746 = 1.228Vac
+20dBu = 7.746Vac
+22dBu = 9.752Vac
+24dBu = 12.28Vac
+4dBu is roughly line level for domestic
+20dBu is roughly line level for PA.
You need mighty high supply voltages to allow the electronics to get up there cleanly.
But many use balanced impedance and that reduces the line to ground voltage by 50%. for the half differential signal.
0dBu = 0.7746Vac
+4dBu = 1.585 * 0.7746 = 1.228Vac
+20dBu = 7.746Vac
+22dBu = 9.752Vac
+24dBu = 12.28Vac
+4dBu is roughly line level for domestic
+20dBu is roughly line level for PA.
You need mighty high supply voltages to allow the electronics to get up there cleanly.
But many use balanced impedance and that reduces the line to ground voltage by 50%. for the half differential signal.
+4dBu professional audio gear recording and PA and -10dBV consumer or home entertainment audio equipment.
for example;
-10dBV = 0.316 volts AC
0.316 volts AC = -7.79dBu
-10dBV = -7.79dBu
+4dBu - (-7.79dBu) = 11.79dB difference
in other words, a -10dBv signal is 11.79dB lower than +4dBu
or for example;
+4dBu = 1.23 volts AC
1.23 volts AC = 1.79dBV
+4dBu = 1.79dBV
1.79dBV - (-10dBV) = 11.79dB difference
in other words, a +4dBu signal is 11.79dB greater than -10dBV
for example;
-10dBV = 0.316 volts AC
0.316 volts AC = -7.79dBu
-10dBV = -7.79dBu
+4dBu - (-7.79dBu) = 11.79dB difference
in other words, a -10dBv signal is 11.79dB lower than +4dBu
or for example;
+4dBu = 1.23 volts AC
1.23 volts AC = 1.79dBV
+4dBu = 1.79dBV
1.79dBV - (-10dBV) = 11.79dB difference
in other words, a +4dBu signal is 11.79dB greater than -10dBV
Last edited:
have you seen these.
Rolls Corporation - Real Sound - Products MX41b Four Channel Mixer
Rolls Corporation - Real Sound - Products MX42 Stereo Mini Mixer
take note that the output level is specified with a (-) negative value
Rolls Corporation - Real Sound - Products MX41b Four Channel Mixer
Rolls Corporation - Real Sound - Products MX42 Stereo Mini Mixer
take note that the output level is specified with a (-) negative value
So, if I am understanding this correctly:
I am using op amps for the following reasons:
1) To increase the signal gain so that any voltage drops due to the circuitry of my mixer and amplified so that the output signal voltage ultimately matches the input signal.
2) As a matching circuit so that I can have a high impedance at the input and low impedance at the output.
3) As a comparator for the LED circuit so it lights up at a particular input level.
Anything else I missed?
I am using op amps for the following reasons:
1) To increase the signal gain so that any voltage drops due to the circuitry of my mixer and amplified so that the output signal voltage ultimately matches the input signal.
2) As a matching circuit so that I can have a high impedance at the input and low impedance at the output.
3) As a comparator for the LED circuit so it lights up at a particular input level.
Anything else I missed?
I thought it would be a resistor - so as to show high impedance for the input signal.
Look at your input circuit. The input impedance is typically determined by a resistor to ground in the single ended non inverting case. Balanced impedance connections will have different arrangements unless an instrumentation amp setup is used.
There have been many threads on what is the best opamp. That all depends on the specific application and performance goals. While you are new to this and playing with breadboards I suggest sticking with something like a TL-072 which will give ok audio performance without being too fussy about circuit layout or having to worry about input currents. A step up would be an OPA-2134 which sound pretty darned good but are still ok with beginner circuit layouts.
There have been many threads on what is the best opamp. That all depends on the specific application and performance goals. While you are new to this and playing with breadboards I suggest sticking with something like a TL-072 which will give ok audio performance without being too fussy about circuit layout or having to worry about input currents. A step up would be an OPA-2134 which sound pretty darned good but are still ok with beginner circuit layouts.
Thanks Bob, I'll look into those Op Amps.
In the meantime, I was looking at the specs of the Pioneer DJM900Nexus DJ mixer which is a standard in clubs and bars.
On the output, it has "output level/Load impedance/Output impedance" as the following:
+8 dBu/10kΩ/5Ω or lower
What's the difference between the "load impedance" and "output impedance"? What will my mixer see - in terms of impedance - if it goes after this mixer in the signal path?
https://www.pioneerelectronics.com/.../DJM-900nexus_OperatingInstructions070511.pdf
In the meantime, I was looking at the specs of the Pioneer DJM900Nexus DJ mixer which is a standard in clubs and bars.
On the output, it has "output level/Load impedance/Output impedance" as the following:
+8 dBu/10kΩ/5Ω or lower
What's the difference between the "load impedance" and "output impedance"? What will my mixer see - in terms of impedance - if it goes after this mixer in the signal path?
https://www.pioneerelectronics.com/.../DJM-900nexus_OperatingInstructions070511.pdf
I'm 6 days post knee replacement so hopefully someone will get you back on track if I say something goofy.
As for the signal level, Turk posted on that, so you can determine the voltage you will send to your mixer.
Load impedance is what the mixer would like to see as the input impedance of the following device. Since you're designing the next stage, you can set its input impedance to 10KOhms.
Output impedance is a measure of how the output deviates from being a perfect voltage source. It's conceptually like putting a resistor of that value in series wit the output. The voltage delivered to the input of the following stage is reduced with reducing input impedance. Let's say you have a tube microphone preamp with an output impedance of 4,500 ohms that outputs 6V into an infinite load. Plug that into a mixer with a 10K input impedance and what happens to the input voltage? Calculate as a resistive voltage divider with 6V into a 4.5K and 10K resistor in series. Solve for voltage across the 10K. I'll wait...
You should have come up with something close to 4V. This is a rather extreme example of why you want the input impedance much much higher than the output impedance of the source driving it.
Another issue with high output impedance already mentioned is its interaction with cable capacitance and the next stage's input capacitance.
The stated 5 ohm output impedance means you won't lose much signal strength or bandwidth with reasonable interconnects to your project mixer. The output stage may not have great current capability or distortion rises significantly with current. Keep it happy with a 10-20K load.
I don't think you've answered why you want to build a mixer ehen you have one. Do you need additional input(s)?
Hope I'm somewhat coherent give the pain medicine I've taken to make PT tolerable.
As for the signal level, Turk posted on that, so you can determine the voltage you will send to your mixer.
Load impedance is what the mixer would like to see as the input impedance of the following device. Since you're designing the next stage, you can set its input impedance to 10KOhms.
Output impedance is a measure of how the output deviates from being a perfect voltage source. It's conceptually like putting a resistor of that value in series wit the output. The voltage delivered to the input of the following stage is reduced with reducing input impedance. Let's say you have a tube microphone preamp with an output impedance of 4,500 ohms that outputs 6V into an infinite load. Plug that into a mixer with a 10K input impedance and what happens to the input voltage? Calculate as a resistive voltage divider with 6V into a 4.5K and 10K resistor in series. Solve for voltage across the 10K. I'll wait...
You should have come up with something close to 4V. This is a rather extreme example of why you want the input impedance much much higher than the output impedance of the source driving it.
Another issue with high output impedance already mentioned is its interaction with cable capacitance and the next stage's input capacitance.
The stated 5 ohm output impedance means you won't lose much signal strength or bandwidth with reasonable interconnects to your project mixer. The output stage may not have great current capability or distortion rises significantly with current. Keep it happy with a 10-20K load.
I don't think you've answered why you want to build a mixer ehen you have one. Do you need additional input(s)?
Hope I'm somewhat coherent give the pain medicine I've taken to make PT tolerable.
OOooooooo - sounds tough. Hang in there pal. 🙂
So they're essentially giving a suggestions as to what would be a good load? So surely if I make it higher (in accordance with impedance bridging theory - YES - I am starting to learn some stuff now - hehehe), it will be even better? I noticed a lot of balanced input modules use 47k resistors. Is that common? Why 47k?
Right...because of the voltage drop when you add the new resistance - got that. The exact figure is 4.137931034 Volts. Hehe - I'm learning here. 🙂
Am I also correct in saying that the amperage drops from I = 1.333333 mA to I = .413793103 mA by adding the new load?
That is a lot of loss. It should be possible to then boost that voltage back up again with an Op Amp right? Question is: Does using this method introduce unwanted noise and a decreased signal-to-noise ratio?(this is just an aside - which I thought was interesting).
Call it a suggested load, call it defining the conditions under which it will give you the specified output voltage. Doesn't really matter.
Yes, you'd end up with about that much output current, but first case was assuming a disconnected or infinite impedance load. Your high current case describes a shorted load. Remember that 4K5 resistor in series with a perfect voltage source emulates a real world imperfect voltage source. It doesn't really exist.
Your noise question is more apropos than I think you realize. That's the tradeoff in designing your input impedance. On one hand higher impedance implies more resistor noise, on the other hand go too low and your source may have trouble driving it. I don't know the design criteria that lead to so many 47K input impedances. I have a theory that it dates to the dawn of the transistor age as a compromise between High enough for tube sources that were once plentiful and low enough for early transistors/op amps (Some current must flow in the input resistor to bias that transistor. Higher resistances mean higher voltages resulting in DC offset.) In modern solid state gear 10-20K input impedance is becoming more common. Musicians often use tube gear, which may be why mixers are maintaining high input impedance. Why 47K and not a nice round 50? Check the standard values for 10% resistors which were about all you could reasonably afford in the 70s.
In your case, I see no reason to use more than 10K input impedance. Your mixer and iPod can both drive it easily.
Capacitors in series with the inputs are there to block DC that might be on the output of your source. They form a high pass filter with the input impedance.
Capacitors between the signal and ground are there to form a low pass RC filter with the series resistance that bandwidth limits the stage (amp, mixer etc.)
I've attached a Leach amp schematic to illustrate this. With the input on pins 2 and 1, C6 and C7 are the DC blocking caps. They form a high pass filter with R17 (which also sets the input impedance). R11 and C8 form the low pass bandwidth limiting filter.
You seem to be getting the RC filter concept. If you hadn't already figured it out, the response of such a filter is not a sharp corner. If you put a 150KHz low pass on the input it does not mean that signals up to that point are unchanged and those above are completely attenuated. I suggest that you either find an online calculator that plots the response of the filter or download a circuit simulation program that will allow you to plug in the values appropriate for your equipment and see what happens to the output.
Knee replacement isn't as bad as I feared. Yes, there's pain, but a week out I'd put baseline pain about the same as what I had on a day to day basis before surgery. I seem to be recovering more quickly than most, but went into it with my legs in good shape. Sounds crazy, but I couldn't walk much even with a cane and I rode my bike 15 miles 4 days before surgery and averaged around 55 miles a week for the last few months.
Yes, you'd end up with about that much output current, but first case was assuming a disconnected or infinite impedance load. Your high current case describes a shorted load. Remember that 4K5 resistor in series with a perfect voltage source emulates a real world imperfect voltage source. It doesn't really exist.
Your noise question is more apropos than I think you realize. That's the tradeoff in designing your input impedance. On one hand higher impedance implies more resistor noise, on the other hand go too low and your source may have trouble driving it. I don't know the design criteria that lead to so many 47K input impedances. I have a theory that it dates to the dawn of the transistor age as a compromise between High enough for tube sources that were once plentiful and low enough for early transistors/op amps (Some current must flow in the input resistor to bias that transistor. Higher resistances mean higher voltages resulting in DC offset.) In modern solid state gear 10-20K input impedance is becoming more common. Musicians often use tube gear, which may be why mixers are maintaining high input impedance. Why 47K and not a nice round 50? Check the standard values for 10% resistors which were about all you could reasonably afford in the 70s.
In your case, I see no reason to use more than 10K input impedance. Your mixer and iPod can both drive it easily.
Capacitors in series with the inputs are there to block DC that might be on the output of your source. They form a high pass filter with the input impedance.
Capacitors between the signal and ground are there to form a low pass RC filter with the series resistance that bandwidth limits the stage (amp, mixer etc.)
I've attached a Leach amp schematic to illustrate this. With the input on pins 2 and 1, C6 and C7 are the DC blocking caps. They form a high pass filter with R17 (which also sets the input impedance). R11 and C8 form the low pass bandwidth limiting filter.
You seem to be getting the RC filter concept. If you hadn't already figured it out, the response of such a filter is not a sharp corner. If you put a 150KHz low pass on the input it does not mean that signals up to that point are unchanged and those above are completely attenuated. I suggest that you either find an online calculator that plots the response of the filter or download a circuit simulation program that will allow you to plug in the values appropriate for your equipment and see what happens to the output.
Knee replacement isn't as bad as I feared. Yes, there's pain, but a week out I'd put baseline pain about the same as what I had on a day to day basis before surgery. I seem to be recovering more quickly than most, but went into it with my legs in good shape. Sounds crazy, but I couldn't walk much even with a cane and I rode my bike 15 miles 4 days before surgery and averaged around 55 miles a week for the last few months.
Last edited:
In the leach amp schematic: why are caps C6 & C7 only on pin 2 and not pin 3? Surely one would want to remove the D.C. Offset from both pins of a balanced input?
Sent from my iPhone using Tapatalk
Sent from my iPhone using Tapatalk
My next question is about headroom on the input: the pioneer states +8dBu. Should I allow some headroom above that? How much? What's the norm?
Sent from my iPhone using Tapatalk
Sent from my iPhone using Tapatalk
Right number of pins, wrong input configuration. The third pin allows a DC connection if you want that. There are caps in the feedback circuit that reduces the gain to 1 so you are less likely to blow out your speakers with a failure.
There are those that don't hear a difference between capacitors. Some hear a difference. I'm in the second camp. Not to argue it here, why not get rid of as many coupling capacitors as possible? If the output of the source is cspacitively coupled you can safely use the direct connection.
There are those that don't hear a difference between capacitors. Some hear a difference. I'm in the second camp. Not to argue it here, why not get rid of as many coupling capacitors as possible? If the output of the source is cspacitively coupled you can safely use the direct connection.
- Status
- Not open for further replies.