DIY Project #03: Buffered Multiples
Jul 30, 2019 12:12:07 GMT
admin, robertlanger, and 6 more like this
Post by NightMachines on Jul 30, 2019 12:12:07 GMT
!!! WARNING / DISCLAIMER !!!
Everything you do as a result from reading the following posts, you do so at your own risk. While synth DIY is generally safe, you are working with electricity, hot soldering irons, etc. and those things can be dangerous to your music equipment, house and to your health. Please always concentrate when working on your DIY projects and be calm and careful.
DIY Project #03: Buffered Multiples
Awesome! You’re still reading! Then get ready for your first active DIY module. So far we’ve only built passive circuits, meaning that they don’t require power to function. Our attenuator pots and OR logic diodes simply received an input signal and processed it, without the components themselves being connected to a power supply. Thats very cool and also very safe. However we can only do so much with passive components ... well, actually there is a lot you can do with passive circuits, but sooner or later active components will come in very handy, so let us build a very simple active circuit now, which is also useful in your modular synth: a buffered multiple!
You are already familiar with passive multiples in your AE Modular system and also as rows on your breadboard.
Passive multiples are simply sockets which are directly connected to copy an incoming signal to several outputs.
That’s great and most of the time passive multiples work perfectly for that purpose. Only when they need to copy an incoming signal precisely, for example when we want to copy exact pitch CV from one sequencer to multiple VCOs, we may run into a problem.
When taking a signal from a complex circuit like a sequencer and simply spreading it out to multiple other complex circuits like VCO modules, each additional signal copy can make the overall signal voltage decrease a little bit, resulting in an out-of-tune melody. In the case of pitch CV, this voltage drop can already be heard after the second copy (try it!), whereas you might not notice it at all on other occasions, like when using the signal copies for filter cutoff or on other modulation targets. It is however almost always there when using passive multiples.
To avoid the voltage drop, we need to decouple all of those complex circuits and create fresh and clean copies of our signal. This process is called buffering and we will use an MCP602 Operational Amplifier (Op-Amp) to achieve it.
Op-Amps are very plain looking integrated circuits (ICs), but they can be used for a lot of purposes. For our buffered multiples, we will use them as so-called Voltage Followers, which is also their most simple configuration:
The big triangle is the Op-Amp symbol. It is connected to +5V and GND so that the IC receives power in order to do its thing. Then it has a non-inverting input (+), an inverting input (-) and an output (right tip of the triangle). Our input signal which we want to buffer goes into the non-inverting input and from the Op-Amp’s output there is a feedback loop back to its inverting input, as well as a connection to our module’s output socket. This circuit will make the Op-Amp simply measure the input voltage and then instantly recreate it with fresh electricity from the +5V power supply. So our input signal path actually stops at the Op-Amp’s input completely and a newly born clone of it is then sent onwards. Creepy, but useful
Buffers like this are everywhere in electronics. If we didn’t have them, then all of those AE modules patched together would create one big complex circuit, where turning a potentiometer on one module could affect a potentiometer value on another module or even worse, make the whole synthesizer unstable und unusable. So it's a good idea to keep modules (or even parts of modules) "buffered" from each other, often done by adding an Op-Amp voltage follower on their outputs.
But that black Op-Amp IC looks very different from the triangle symbol in the schematic and there are even more metal legs on the IC than connections on the symbol. What's up with that? Well, ICs usually all look like that black box, varying mostly just in length and the amount of legs.
A thru-hole IC with two parallel lines of connectors like our MCP602 above is called a "Dual In-line Package" (DIP) or "Plastic Dual In-Line Package" (PDIP). In order to know which purpose each metal leg serves, we need to look into the IC’s datasheet. Let's do that now ... but beware, datasheets are not for the faint of heart! Open the following link.
MCP602 DATASHEET PDF
Oh gee, that's a lot of pages and graphs and stuff ... luckily, we only need to look for a few things on the very first page:
"Single supply" means that the Op-Amp only requires one power supply and in the MCP602's case, our AE Modular +5V bus fits very well. A "dual supply" Op-Amp usually requires a negative voltage power supply as well, but we don't have that in our AE Modular.
"Rail-to-Rail" means that the Op-Amp can output voltages very close to the power supply rail ratings it is connected to, in our case GND and +5V. That's not normal for all Op-Amps and it can make them more expensive. Because our AE Modular power supply is the same range as the signals we use in the synth though, we really want a rail-to-rail Op-Amp to ideally process signals over the full range between 0V/GND and +5V. The MCP602 cannot reach full +5V in its output, but it gets close enough for the purpose of this beginner’s DIY project.
"Input Range Includes Ground" is also very nice to have, because otherwise we might need to create a new virtual ground for our Op-Amp, which would be more complex.
Reading on:
Great! Here we find the so-called pinout of our MCP602 PDIP IC, describing what each metal leg is used for. The legs or pins are numbered counter-clockwise from 1 to 8 and there is a notch to help us with the orientation.
VDD: The positive power supply pin, in our case where we connect the +5V bus
VSS: The negative or "low" power supply pin, where we connect GND
Then we have A and B pins with the same functions, because there are actually two separate Op-Amps in our tiny little IC! Neat!
VIN-: inverting input
VIN+: non-inverting input
VOUT: output
Easy! Here's a side-by-side image of our schematic, the left side of the IC and the pinout:
With all of this understood, let's draw a circuit diagram for a buffered multiple with one input and four outputs, using two MCP602 Op-Amp ICs.
Oh, that doesn't look too difficult! On the left we have our input socket, which is wired to four Op-Amp non-inverting (+) inputs, which simply measure the incoming voltage. Due to their voltage follower configuration (feedback-loop) the Op-Amps simply recreate the input voltage on their outputs, which we wire to our output sockets.
On the bottom left there is also a 100k Ohm (or higher) pull down resistor, which makes sure that the Op-Amp inputs always receive a clearly defined voltage. Either from the input socket or, if nothing is patched in there, simply a "low" voltage from GND. This keeps the Op-Amps stable when nothing is patched and they won't be floating. It's the same concept we encountered in our OR logic DIY project.
Pulling inputs to ground doesn't work with all Op-Amps, but since we know from the MCP602's datasheet that its "input range includes ground" we can do that here. On Op-Amps where this is not the case, we would need to feed the unconnected inputs a higher voltage, which would be more complicated.
Now, on to the breadboard!!!
Breadboard Circuit Test
Bill of materials (BOM):
1x MCP602 Op-Amp (AliExpress)
1x 100k Ohm Resistor (or higher, up to 1M Ohm)
Setup
!!! ATTENTION !!!
Switch off the power to your AE Modular system before you start!
To test our circuit on the breadboard, let's only use one MCP602 IC, which houses two Op-Amps, so we can get two copies of our input signal.
Here's how it might look, with a bunch of colorful annotations:
... and here is the clean photo, if that's easier to follow:
Note how that groove in the middle of the breadboard was made perfectly for DIP ICs, keeping their parallel lines of pins separate.
Please make sure that you connect the +5V and GND wires to the correct sockets on your AE Modular bus connector. Here's that picture from the last projects again:
!!! ATTENTION !!!
Double-check your patch cables now! You will switch on the power to the system next. This means that your DIY circuit is connected to your modular synth and if you made a mistake, things might break. After powering your system on, immediately check if all things look normal. If you feel something is behaving weirdly (e.g. no LEDs light up, no sound, weird smell) disconnect the power immediately and re-check your cabling.
Now, turn on the power to your AE Modular system and patch a pitch CV signal into the input socket on the breadboard and connect the two "COPY" outputs to two VCO pitch CV inputs and listen to their frequencies. Then try the same with a passive multiple and you will notice the pitch being slightly lower as soon as you connect the second VCO, because of the voltage drop.
Congratulations! You just built your first active synth DIY circuit and wow ... over the course of three DIY projects you've learned a lot of important basics about synth DIY and electronics in general. With this knowledge, future DIY projects will become easier to understand while offering even more exciting features for your AE Modular synth than those bread and butter modules we've been building thus far.
What next?
Next you could try to add a second MCP602 IC to the breadboard to create a total of four copies of the signal, just like we planned it in our schematic earlier. Once you feel comfortable with all of this. You can take the parts off the breadboard, solder them onto a PCB and build your own AE Modular Buffered Multiples module. In fact, this is what I will show you next in this thread. So stay tuned
Everything you do as a result from reading the following posts, you do so at your own risk. While synth DIY is generally safe, you are working with electricity, hot soldering irons, etc. and those things can be dangerous to your music equipment, house and to your health. Please always concentrate when working on your DIY projects and be calm and careful.
DIY Project #03: Buffered Multiples
Awesome! You’re still reading! Then get ready for your first active DIY module. So far we’ve only built passive circuits, meaning that they don’t require power to function. Our attenuator pots and OR logic diodes simply received an input signal and processed it, without the components themselves being connected to a power supply. Thats very cool and also very safe. However we can only do so much with passive components ... well, actually there is a lot you can do with passive circuits, but sooner or later active components will come in very handy, so let us build a very simple active circuit now, which is also useful in your modular synth: a buffered multiple!
You are already familiar with passive multiples in your AE Modular system and also as rows on your breadboard.
Passive multiples are simply sockets which are directly connected to copy an incoming signal to several outputs.
That’s great and most of the time passive multiples work perfectly for that purpose. Only when they need to copy an incoming signal precisely, for example when we want to copy exact pitch CV from one sequencer to multiple VCOs, we may run into a problem.
When taking a signal from a complex circuit like a sequencer and simply spreading it out to multiple other complex circuits like VCO modules, each additional signal copy can make the overall signal voltage decrease a little bit, resulting in an out-of-tune melody. In the case of pitch CV, this voltage drop can already be heard after the second copy (try it!), whereas you might not notice it at all on other occasions, like when using the signal copies for filter cutoff or on other modulation targets. It is however almost always there when using passive multiples.
To avoid the voltage drop, we need to decouple all of those complex circuits and create fresh and clean copies of our signal. This process is called buffering and we will use an MCP602 Operational Amplifier (Op-Amp) to achieve it.
Op-Amps are very plain looking integrated circuits (ICs), but they can be used for a lot of purposes. For our buffered multiples, we will use them as so-called Voltage Followers, which is also their most simple configuration:
The big triangle is the Op-Amp symbol. It is connected to +5V and GND so that the IC receives power in order to do its thing. Then it has a non-inverting input (+), an inverting input (-) and an output (right tip of the triangle). Our input signal which we want to buffer goes into the non-inverting input and from the Op-Amp’s output there is a feedback loop back to its inverting input, as well as a connection to our module’s output socket. This circuit will make the Op-Amp simply measure the input voltage and then instantly recreate it with fresh electricity from the +5V power supply. So our input signal path actually stops at the Op-Amp’s input completely and a newly born clone of it is then sent onwards. Creepy, but useful
Buffers like this are everywhere in electronics. If we didn’t have them, then all of those AE modules patched together would create one big complex circuit, where turning a potentiometer on one module could affect a potentiometer value on another module or even worse, make the whole synthesizer unstable und unusable. So it's a good idea to keep modules (or even parts of modules) "buffered" from each other, often done by adding an Op-Amp voltage follower on their outputs.
But that black Op-Amp IC looks very different from the triangle symbol in the schematic and there are even more metal legs on the IC than connections on the symbol. What's up with that? Well, ICs usually all look like that black box, varying mostly just in length and the amount of legs.
A thru-hole IC with two parallel lines of connectors like our MCP602 above is called a "Dual In-line Package" (DIP) or "Plastic Dual In-Line Package" (PDIP). In order to know which purpose each metal leg serves, we need to look into the IC’s datasheet. Let's do that now ... but beware, datasheets are not for the faint of heart! Open the following link.
MCP602 DATASHEET PDF Oh gee, that's a lot of pages and graphs and stuff ... luckily, we only need to look for a few things on the very first page:
"Single supply" means that the Op-Amp only requires one power supply and in the MCP602's case, our AE Modular +5V bus fits very well. A "dual supply" Op-Amp usually requires a negative voltage power supply as well, but we don't have that in our AE Modular.
"Rail-to-Rail" means that the Op-Amp can output voltages very close to the power supply rail ratings it is connected to, in our case GND and +5V. That's not normal for all Op-Amps and it can make them more expensive. Because our AE Modular power supply is the same range as the signals we use in the synth though, we really want a rail-to-rail Op-Amp to ideally process signals over the full range between 0V/GND and +5V. The MCP602 cannot reach full +5V in its output, but it gets close enough for the purpose of this beginner’s DIY project.
"Input Range Includes Ground" is also very nice to have, because otherwise we might need to create a new virtual ground for our Op-Amp, which would be more complex.
Reading on:
Great! Here we find the so-called pinout of our MCP602 PDIP IC, describing what each metal leg is used for. The legs or pins are numbered counter-clockwise from 1 to 8 and there is a notch to help us with the orientation.
VDD: The positive power supply pin, in our case where we connect the +5V bus
VSS: The negative or "low" power supply pin, where we connect GND
Then we have A and B pins with the same functions, because there are actually two separate Op-Amps in our tiny little IC! Neat!
VIN-: inverting input
VIN+: non-inverting input
VOUT: output
Easy! Here's a side-by-side image of our schematic, the left side of the IC and the pinout:
With all of this understood, let's draw a circuit diagram for a buffered multiple with one input and four outputs, using two MCP602 Op-Amp ICs.
Oh, that doesn't look too difficult! On the left we have our input socket, which is wired to four Op-Amp non-inverting (+) inputs, which simply measure the incoming voltage. Due to their voltage follower configuration (feedback-loop) the Op-Amps simply recreate the input voltage on their outputs, which we wire to our output sockets.
On the bottom left there is also a 100k Ohm (or higher) pull down resistor, which makes sure that the Op-Amp inputs always receive a clearly defined voltage. Either from the input socket or, if nothing is patched in there, simply a "low" voltage from GND. This keeps the Op-Amps stable when nothing is patched and they won't be floating. It's the same concept we encountered in our OR logic DIY project.
Pulling inputs to ground doesn't work with all Op-Amps, but since we know from the MCP602's datasheet that its "input range includes ground" we can do that here. On Op-Amps where this is not the case, we would need to feed the unconnected inputs a higher voltage, which would be more complicated.
Now, on to the breadboard!!!
Breadboard Circuit Test
Bill of materials (BOM):
1x MCP602 Op-Amp (AliExpress)
1x 100k Ohm Resistor (or higher, up to 1M Ohm)
Setup
!!! ATTENTION !!!
Switch off the power to your AE Modular system before you start!
To test our circuit on the breadboard, let's only use one MCP602 IC, which houses two Op-Amps, so we can get two copies of our input signal.
Here's how it might look, with a bunch of colorful annotations:
... and here is the clean photo, if that's easier to follow:
Note how that groove in the middle of the breadboard was made perfectly for DIP ICs, keeping their parallel lines of pins separate.
Please make sure that you connect the +5V and GND wires to the correct sockets on your AE Modular bus connector. Here's that picture from the last projects again:
!!! ATTENTION !!!
Double-check your patch cables now! You will switch on the power to the system next. This means that your DIY circuit is connected to your modular synth and if you made a mistake, things might break. After powering your system on, immediately check if all things look normal. If you feel something is behaving weirdly (e.g. no LEDs light up, no sound, weird smell) disconnect the power immediately and re-check your cabling.
Now, turn on the power to your AE Modular system and patch a pitch CV signal into the input socket on the breadboard and connect the two "COPY" outputs to two VCO pitch CV inputs and listen to their frequencies. Then try the same with a passive multiple and you will notice the pitch being slightly lower as soon as you connect the second VCO, because of the voltage drop.
Congratulations! You just built your first active synth DIY circuit and wow ... over the course of three DIY projects you've learned a lot of important basics about synth DIY and electronics in general. With this knowledge, future DIY projects will become easier to understand while offering even more exciting features for your AE Modular synth than those bread and butter modules we've been building thus far.
What next?
Next you could try to add a second MCP602 IC to the breadboard to create a total of four copies of the signal, just like we planned it in our schematic earlier. Once you feel comfortable with all of this. You can take the parts off the breadboard, solder them onto a PCB and build your own AE Modular Buffered Multiples module. In fact, this is what I will show you next in this thread. So stay tuned
maybe a mistake but gnarly through-hole projects always take up more space than nice neat modern SMD boards.
