Monday, 9 December 2019

Synthesizer Build part-8: 8 STEP SEQUENCER.

A simple 'Baby 8' type Sequencer made with the CD4017 chip. Easy to build and fun to use. No DIY synth should be without one of these.

There are PCBs available for this sequencer. See 'PCB Service'

EDIT: There is now a new and improved version of this sequencer available on this website. I redid the design and included some extra features like external clock input and a CV Offset control. This makes the sequencer much better to use and it is no more complicated to build than this original design. So please go over to project 49: 8 step sequencer version 2 if you want to build this 8 step sequencer.

This sequencer is one of my earlier projects and of my own design although it's more or less put together from bits of other designs like the 'Baby 8' but it works fine for me and is really easy to make and easy to tune although to build it is quite time consuming and repetitive work because a lot of steps have to be soldered eight times. I found it rather tedious work but very worth while. 

A NOTE FOR BEGINNERS: A sequencer does not actually produce any sound itself. It produces a stepped control voltage that can be routed into the CV input of a Voltage Controlled Oscillator and the VCO then produces the actual notes you hear. In a sequencer you can set each of these eight steps or notes manually (with a potmeter for each step) to any voltage/note you want.

Here's the schematic drawing for this sequencer. The connections of the rotary switch are not correct in the schematic. They must be offset by one step from those of the potmeters. So step 1 is reset by the pulse from step 2 so pin one of the switch goes to output 2 of the CD4017, pin 2 of the switch goes to output 3 of the CD4017 etc, etc.


In the schematic above the on/off switch is placed after the voltage regulator to easily switch the sequencer on or off without causing switching pulses on the voltage rails. The complete on/off switching is done with the switch of the powersupply which controls the power of your whole modular synthesizer.

Here's the stripboard layout I made for the sequencer. In the schematic I drew in switches that you can add to turn individual channels on or off but I didn't include them in my build because I didn't have the space for them on the panel. In this layout I don't use any transistors either. I thought it was nonsense to make this more difficult then it needs to be. It will work fine without them because we hardly draw any current from these outputs. The CV output signal goes straight into a VCO. The layout has an extra 10µF electrolytic capacitor on the output of the voltage regulator that is not on the schematic. It's for extra noise suppression. You can get away with using a 100nF cap or leaving it out completely.
Be careful when you wire this up, note that the jumper (or wire bridge) for output 5 is connected to pin 10 of the chip so the left bunch of jumpers skips a copper trace at output 5. Look carefully at the layout! If you want to include switches to mute individual channels then put them in series with the diode!

(Last revised: 26-Feb.-2020: Minor cosmetic changes.)

NOTE: All potmeters in the layout are shown from the front side!

Use Schottky Diodes on the wipers of the potmeters. They only have a voltage drop of 0.2V instead of the 0.6 to 0.7 Voltage drop over 1N4148 diodes usually found in sequencers like this. This means you can get deeper tones from the VCO you plug it into. Because of the 0.6 to 0.7 Volt voltage drop over the silicone diodes, the first section of the potmeters wouldn't do anything until you get above 0.6 volts. So with a lower voltage drop there's more throw on the potmeter. As an experiment I also installed a 100K potmeter over the output of the Control Voltage and the wiper goes to the CV output jack. That way you can get even lower tones although, of course, this compresses the dynamic range of the sequencer. With the potmeter fully open you get the normal range of 0.2 to 8 Volts. If you close the pot half way, your range becomes 0.1 to 4 Volts so the spacing between notes becomes smaller. You don't need to include that option, I never use it and it is not included in the layout. But anyway, this is an expirimental sequencer and as a whole it works really well, If you build it you will be happy, I guarantee it. :)
A better solution, and one you should consider if you are comfortable with designing simple circuits with opamps, is to add a DC-Offset feature to this sequencer. That way you can get the lowest notes down to 0 volt without influencing the dynamic range of the sequencer. It's easy enough to do. This is not included in the layout or schematic though.

Here's a close-up of the stripboard:



Bill of materials for the layout version. You'll need ten (10) 100K potmeters instead of the 8 mentioned in the B.O.M. below. You need one for speed control and one for offset (if you build version 2 of this sequencer which I strongly advise you to do. Go to project 49) :


Here's a picture of the sequencer:



The sequencer is build up around the CD4017 decade counter chip, using a CD40106 to create the clock pulses which also serve as the 'Gate' pulses.
The CD40106 hex inverter is used as a low frequency oscillator giving off squarewave pulses who's frequency can be controlled by the 100K potmeter. I used a 15µF electrolythic Capacitor although a 10µF will do just as well. But a little higher value will give you slower speeds so you could even try a 22µF cap. The clock pulses can be interrupted by switch S-2 to give you a chance to tune that particular channel. Sometimes it can happen that after using the 'Stop/Run' switch that the sequencer jumps to channel one. If that happens try using a different CD40106 chip. You might have a fake one and they can be quircky in their behaviour.
With S-2 closed the clock pulses go into pin 14 of the CD4017 and with every pulse the chip will output a high signal on a different pin. The order by which the different pins go high is a bit random. Here is the right order: 3,2,4,7,10,1,5,6,9,11. Because of this confusing order, the outputs are set in the right order by the wire bridges to the copper traces underneath the CD4017. From there the pulses can be accessed in the right order to avoid confusion. Following the schematic drawing, the pulses go straight into the base of the 2N2222 transistors which are used here as switches. The Base-Emitter voltage is way more than needed to saturate the transistor and fully open it up. I chose the 2N2222 transistor because it can handle a reasonably large current and there's no need to use any resistors to connect them (although using a resistor in series with the base connection wouldn't be a bad thing because we're using the 2N2222 at near the limit of the operational specs.) From this base connection we also feed the eight LED's which indicate which channel is on at each moment in time. The LED's are connected with 3K resistors to reduce current flow and still provide a bright light.
All the collectors of the transistors are connected straight to the 8 Volt power rail and the emitters are all connected to ground.
It's better to just follow the stripboard layout and skip the whole transistor setup and connect the output of the CD4017 straight to the potmeters. I'm using transistors as a sort of buffer and to make this sequencer future proof for other experiments so I can draw some current from the outputs should that be necessary. But you can just leave them out it you want to. Makes it so much easier.
By setting the different potmeters, you can create the different tonal paterns the sequencer produces.
Because the potmeters are simply used as voltage deviders, it doesn't really matter which value they are as long as it's 50K or over so that they don't draw too much current and as long as you use the same value on all 8 channels.
You can tap the 'Gate' pulses straight from pin 3 of the Speed Control potmeter to the Gate output jack mounted in the panel. The pulses are really clean looking 8 Volt squarewave pulses with a 50% duty cycle so if you use the gate output into the ADSR, it will sound as if a key is pressed every time the sequencer switches to an other note.

A ten step switch is used to select the length of the sequence. It can be anything from 1 to 8. Btw, you can easily make this a ten step sequencer by connecting the last two pins from the CD4017. I made it an 8 step because I didn't have enough space to mount everything horizontally and because 8 steps is more natural for music than 10 steps because you normally have 4 notes in a beat. So multiples of 4 are better. The potmeters on my panel are mounted vertically and I could only fit eight of them below eachother anyway.
Connect the wiper part of the switch to pin 15 of the CD4017 and the wires from 1 to 8 to their relative position on the switch. Connect pins 9 and 10 of the switch together and connect the ninth output from the CD4017 to that. The pulse going into pin 15 of the 4017 will reset the chip and the counter will start over again.
Don't forget to connect pin 13 of the CD4017 to ground.

It is best with this build to make the panel first and connect all the components and do the essential wiring while you have access. Then make the circuitboard and connect the wires to the panel. Solder the resistors straight to the LED's and the diodes to the wipers of the potmeters. Connect the cathodes together and solder a wire from there to the CV output jack.
I used 5mm LED's and I made the holes in the panel by using a drill rather than a hole enlarger bit which I normally use to enlarge the pilot holes I drilled. The drill is usually a little bit less then 5mm and therefor the LED's will sit very tight and don't even need to be glued in place (although it is best to hot-glue them in place anyway).

Do not forget to solder a big 470µF capacitor on the input of the 7808 voltage regulator. Otherwise pulses will bleed through onto the power supply rails and you'll hear the tone sequence even if the sequencer isn't connected to the CV input of the VCO. I also included an ON/OFF switch (S-1) on the panel just to have the option to shut it down. It's the only panel in my synth build to have an ON/OFF switch.

TUNING THE SEQUENCER:
To tune the sequencer, simply set it to the lowest speed and use switch S-2 to interrupt the clock pulses and stop at each channel. Then you can tune that particular channel using a tuner or simply by ear, by turning the potmeter and then you turn switch S-2 back on. The sequencer flips to the next channel, you turn it off again with S-2 and tune that note, then you flip the switch again and jump to the next channel, etc, etc. It's very simple and very effective. :)
A note for beginners. You must connect the CV OUT of the sequencer to the 1V/Oct input of a Voltage Controlled Oscillator (VCO) and the oscillator makes the actual sound. The sequencer only produces a sequence of stepped voltages that the VCO turns into notes so for tuning the sequencer you must have it connected to a VCO.
Because the sequencer can produce any voltage between 1 and 8 Volt it's difficult to set it accurately to a specific note without using a tuner. That's why most professional sequencers have a built in Quantizer which automates this proces. A Quantizer reads an incoming voltage and turns it into the nearest 1/12th of a volt, that way making sure it's a pure note. 
Because most synthesizers use the 1 volt per Octave system and there are 12 notes in an Octave, each note is produced by a multiple of 1/12th of a volt. For instance note C1 = 1.000V, note D1 = 1.083V (1 + 1/12th volt), note F3 = 3.333V (3 + 4/12th volt). So the notes progress upwards in steps of 1/12 of a volt. This sequencer does not have a quantizer and because they are quite difficult to build I don't have a quantizer project on my website. You can however buy them for Eurorack systems. In my Eurorack system I have the Doepfer A-156 QNT which costs about €119 and contains 2 quantizers.
If you are good at working with Arduino's you can easily make a quantizer with that. You can program it to turn any incoming voltage into a multiple of 1/12th of a volt.

Momentary switch:
There is a good way to include a momentary switch mentioned in the comments below by 'tamasgal'. The suggestion is to put a resistor and switch in series connected between V+ and ground and then run a resistor and capacitor from the high potential side of the switch to ground and also connect it to one of the left over schmitt-triggers of the CD40106. Then connect the output to pin 15 of the CD4017. That should take care of any bounce in the momentary switch.
In fact, I have implemented this in version 2 of this sequencer (project 49) and it works really well.

That's all there is to say about this. It's one of the most fun panels for the synthesizer but one of the most tedious to build. It cost me 6 hours straight to design and build it but luckily it worked straight away.

Here's a little demo of the sequencer. This was filmed before I put in switch S-2 so I had no option to tune the sequencer at the time of filming. I might make a new video soon:


Okay, that's another one done. I hope you enjoyed it. If you have any questions about this build then  please leave them in the comment section below or in the Facebook Group.

Synthesizer Build part-7: THE MOOG LADDER FILTER.

The iconic Moog Ladder Filter. This version is built with transistors only, not transistor array IC's. This is an early project of mine so please bear that in mind when reading the article.

A note before we begin: I have PCBs available for the Moog Ladder Filter using the CA3046 transistor array chips. Just go to the 'PCB Service' option at the top of the menu for more details.

Since this article was written I have made a new version of this filter, this time including the CA3046 transistor arrays and it works very well and no need to match transistors with that version. If you want to build this filter I would really advise you to use that layout instead of this one (although this one works fine too of course). You can find it in Chapter 39 (click here).

I used the schematic from Yusynth's website.


Before we start. Most people always want to know if it works on 12V. I tested the filter on dual 12V and it works just fine.
In this schematic the top and bottom transistors are applied in the form of a transistor array chip, the CA3046, but I couldn't get hold of that quickly enough and this early in the build I didn't really trust myself to design a layout including those arrays, so I decided to use all transistors and that works just as well. It makes the layout a lot easier. It is always mentioned that you must use matched pairs of transistors for this filter but really, that's a throw-back to the early seventies when transistors were not as consistent and reliable as they are now so if you have transistors from the same batch they will probably be matched well enough but put them through the transistor tester on your multimeter and match them on hfe value. The only place where the transistors must be matched well is on the place in the schematic where they use the CA3046; the top and bottom of the filter and the output on the side. I personally matched all my transistors by using the Transistor Curve Tracer I described in an earlier article on this website.
I built a second ladder filter as a test for the layout below and I used all unmatched transistors. The layout works fine but using unmatched transistors did not turn out well. I could not get the resonance trimmed correctly and there were enormous differences in volume when using the resonance potmeter. I used a squarewave for testing and the top of the squarewave had an angle to it instead of being horizontal. So you must used matched transistors!
This filter has a few quircks that you need to know about but which are normal for this design.
- The Resonance potmeter has only a small area of influence. For most of the throw of the potmeter you will hardly notice anything. This is normal for this design. That's why we need a reversed logarithmic potmeter for Resonance. To stretch out that last bit of the potmeter.
- When the Resonance is fully open, the output volume drops. This too is normal for this filter and even the original Moog ones have this. Yusynth also talks about this on his website.
- If the audio input level is too low you can loose the self-resonance on the bottom parts of a squarewave. (If you use a squarewave as input wave of course). Again, this is a known quirck of this filter type. It needs decent level of audio input.

The build is quite straight forward but you need to be very accurate. The 50K anti-logarithmic potmeter for the Emphasis or Resonance control was an other thing I couldn't get a hold of so I made my own by using a linear potmeter with a 5K resistor between pins 1 and 2. This works very well, In the layout I used a reversed logarithmic potmeter and I show the alternative that I myself used, next to it.
The input level potmeters are 50K logarithmic ones but if you don't have those just use linear ones. They don't even have to be 50K. You can use 100K or 1M or even 10K if that's what you have available. They're just audio input level pots so they act as attenuators or voltage dividers and the value has no impact on the working of the circuit. This goes for all the level potmeters in all the projects on this website unless it is mentioned otherwise on the layout.
The Frequency Cutoff potmeter however must be a 10K!

I made a layout for stripboard including the wiring. I used this layout to build a second filter and it worked straight away so this layout is verified. (All potmeters viewed from the front):


(Last revised: 24-June-2020: Corrected polarity of C3. 15-July-2020 added alternative for reversed potmeter.)

Stripboard only. Beware that some stripboards are sold with 56 instead of 55 holes horizontally. The layout is 55 holes wide:



Here are a few pictures of the finished circuitboard:



As you can see in the pictures, I added two trimpotmeters which are not on the stripboard layout above. These are two 200K trim pots and they go over pins 1 and 2 of each opamp, to make the gain adjustable. It says in the schematic to 'adjust the value of the feedback resistors according to audio level'. These trimpots make that possible without having to use the soldering iron. It's a bit awkward with the wires but I had to put the potmeters on the print where there was room enough to accommodate them and the wires. Plus I added them as an after thought, so after I made the layout. At least they are all neatly in a row. :-)
For clarity I made a second layout which includes these alterations. If you decide to replicate this then don't forget to remove (or not solder in) the original resistors over pins 1 and 2; the 56K on IC-1 and the 120K on IC-2 because these are replaced by the trimmers, as the layout below shows. Lay-out is verified not only by me but I heard from many people who built this filter successfully using this layout.:


I lowered the Control Voltage input resistors from 100K to 5K6. In the schematic they are 100K but after installing the filter in my synth set-up I noticed that the CV hardly had any effect at all when I connected a LFO to it, so I lowered these to 5K6.

After finishing the build I tested the filter on the oscilloscope first and set all the trimmers to the right positions. It's easy enough to do, you just watch the scope for the best response. Then, after installing the filter you can adjust the trimmers to get the best sound out of it. This filter is self-oscillating, meaning that if you have nothing connected to the inputs and you turn Frequency Cut-off and Resonance all the way up the filter will oscillate of its own. There's no 1V/Oct. input though so resonance won't keep track with the notes on the keyboard. An other frequently occurring problem is that the self resonance only occurs at the top of the squarewave and not the bottom part. When you start it up it will self oscillate on both the top and the bottom but as soon as the transistors warm up, about 20 seconds, the bottom oscillations disappear. This happens when the transistors are not perfectly matched. Using the CA3046 transistor arrays will solve this and that's why I made a second Moog Ladder Filter project using the CA3046 chips. You can find that in chapter 39
The first all transistor filter I built didn't have this problem so if you're careful with matching the transistors you should be fine. Please note that the input volume is also an influence on this. If the level is too low you can also have the bottom oscillations disappear.
Here are some oscilloscope images to illustrate what I mean:
Self oscillation when filter is just switched on:



And here's the situation after about 20 seconds. The bottom oscillations are gone. The filter still sounds pretty cool though:


This is something I only discovered a short while ago but well matched transistors or using the CA3046 chip should solve this issue. The first Ladder Filter I built using all transistors didn't suffer from this problem because I was careful to match the transistors accurately.

MATCHING TRANSISTORS:
 I used to advise people that matching transistors on their Hfe value was good enough but I have found an easy way to do it properly which I advise you to use. This is the Ian Fritz method. We set up a small differential amplifier on a little breadboard and measure the voltage between the two emitters. If the transistors are matched that voltage should be zero. Here's a schematic of the test setup:


The diode is a 1N4148 or any silicon diode will do really. It's voltage drop ensures both transistors get the same base-collector voltage. Make sure the 100K resistors have exactly the same resistance value!
Beware this setup uses a dual 12V powersupply.

VIDEO DEMO:
Here's a video showing the test results on the oscilloscope so you have an idea of what the waveforms should look like. In this first filter I built and which I demo in this video the transistors seem to be reasonably well matched because I do retain some self oscillation on the lower parts of the squarewave:

       


A demonstration of the sound of the filter:



I recently made a Falstad simulation of the ladder filter and it works reasonably well. It shows the resonance and the trimmers work as expected. It even shows the amplitude getting smaller with resonance and self oscillation. You can view the simulation by --- CLICKING HERE --


Here's a look at my PCB for the filter. This is what is actually in my synthesizer and I build it from a different layout that I made earlier especially for the Eurocard format of stripboard:


Here's the panel I made for it:


As you can see in the picture, I've installed a bypass switch for this filter. It's great to have this filter in series with the Korg MS-20 or the ARP2600 filter but sometimes you want to be able to easily switch to one filter. With a bypass switch there's no need to constantly connect and disconnect patch cables so this is really helpful. The switch only works with audio input 1 and it sends the signal straight to the output jack and disconnects the in- and output from the in- and output jacks at the same time. Here's the wiring diagram for the bypass switch:



Okay that's it for this project. Way more synthesizer build articles to find on this website and while you're here leave a comment please!


Wednesday, 4 December 2019

Synthesizer Build part-6: THE SEQUENTIAL PRO ONE LOWPASS FILTER.

The first filter I built and not one of the best in my opinion. Skip to an other project. I'm keeping this up for archival purposes.

[EDIT] Before we start: I'm writing this about 7 months after I published the article below and I can tell you now that I removed this filter from my synth. I couldn't get it to self-oscillate and I was not impressed with the sound of it. Maybe that's due to the fact that I was just starting out with synth building when I made this and I made some mistakes, but if you have the AS3320 chip I advise you to use it for the Digisound 80.6 lowpass filter. That is an amazing sounding filter!
I'm keeping this article up because this website is an archive of all my builds and that includes the ones that were not a succes (That's only this project anyway. The rest all works fine.)

Okay on with the original text:

I've considered many filter designs, and there's a lot to choose from if you search for schematics on the internet, but I've picked out 3 filters that I want in my synthesizer. Two of those are LowPass filters and one has the choise between Lowpass and Highpass. That's the Korg MS20 filter We'll get to that one later.

Now we will concern outselves with the Sequential Circuits Pro One (and many others) Lowpass filter based on the AS3320 chip. I went to the ElectricDruide website and found a page full of filters and amongst them I found this circuit:



There are also versions which are re-configurable with the flick of a switch but they were a bit too complicated for my taste and I thought this would be a great project to start with. Well, it was a good project to learn from but the result was nothing to write home about.

I made a layout on stripboard which you can find in the picture below. All potmeters viewed from the front.:


(Layout revised 15-Feb-2021)

Print only:

Bill of Materials:


The first version I made of this had capacitors that were 220pF instead of the 150pF as seen in this schematic. I thought that would give me more control over the low frequencies and I was right, LOL! It brought the top cut-off frequency down to about 200Hz. Way too low. So I soldered in the 150pF and everything worked after that.

The three inputs in this schematic have different values but I did some testing and the 120K resistor worked the best on the input so I put those in all 3 inputs. I think you should also do this if you decide to build this filter because these schematics reflect the circuit that is used in the synthesizer itself and that's why the inputs have different values because they receive signals with different amplitudes. In our synth all the VCO's produce signals with a set amplitude of 10V peak to peak, therefore the input resistors need to be the same value.

There's not much more to say about this. The opamp in the schematics can be any low noise audio type. I used the TL072 for that because I have a lot of those in my components collection and they are the opamps most used in synthesizer projects. But you can also use the TL082 or NE5532.

You can use this filter with a dual 12V powersupply without any changes except for the current limiting resistor to pin 13. Change it from 1K5 to 1K2 for -12V operation.

This filter sounds like a lowpass filter should although it is not as versatile as the Moog or Korg filters. For one, this filter is not self oscillating (at least, the one I build isn't) so you can't make it scream but it does produce that resonant synthesizer sound, if you know what I mean.  It's just a different filter to the Korg or the Moog Ladder Filter which is an other filter I've put in my synth build. It's more a filter to round off certain sounds you've created on your synth, to make them fuller. I don't know, you have to listen to it to know what I'm on about. It should be capable of self oscillation if you read the text that goes with the schematics. The fact that mine didn't self oscillate is probably due to my inexperience at the time I built this.
EDIT 20-March-2020: I just finished work on the Digisound 80.6 lowpass filter using the AS3320 chip and it sounds fenominal. If I were you I would build that filter instead of this one. 

Here are some pictures of the PCB and the panel I made:



The filter is the one on the right. The other one is the VCO. I had yet to add the text to the panels.



You can see the inputs for oscillators one, two and for noise on the left and below that the resonance control voltage input (for the LFO for instance). That input has a level control and there's a switch for internal or external resonance control. On the right is the audio output. The top potmeter is for the Cut-off frequency and the one below is, as mentioned, for resonance.

Tuesday, 26 November 2019

Synthesizer Build part-5: TIPS FOR SYNTH BUILDING and GENERAL KNOWLEDGE

All sorts of facts you need to know and an bit about how I made my synthesizer cabinets. A glossary explaining terms you hear a lot in the synthesizer world.

My first synth cabinet:
I'm not going to get too deep into this because every individual will make their own cabinet or case to their own taste I think. This cabinet represents the vintage 70's look that I wanted for my synth and I'm very happy with it. 
If you are going to build all, or almost all, of the projects on my website you're going to end up with an amazing synthesizer (that I call the "Bergman-Berlin") that can rival the big Moog systems of back in the day and will cost you far less money. You can build as many oscillators as you wish (my system has five) and a variety of filters and other sound manipulators and noise generators and mixers that will make this an amazing sounding synth. There are examples enough in the demo videos on this site. Here is a picture of my synthesizer as it now is, at the end of October 2020:




Here are some pictures from different stages of building the first cabinet:



Almost finished, just one more panel to fit but already working very well. You can see I incorporated a 'Keyboard Garage' in the case so I can push the keyboard underneath the synth if I'm not using it, to free up space for other things.



I made no drawings and I measured everything as I was building it. That's the way I usually approach woodwork. The drawing is in my head. I did make two cardboard templates for the side panels to make sure I got those exactly the same. I measured the current draw with all panels that I have build so far switched on. All together it drew 250mA. That's less than an old fashioned bicycle lamp. :) I also installed a temperature sensor that is directly in contact with the heatsink of the LM317 that regulates the 15 volt output and it runs up to about 60°C. That's perfectly fine and normal. It can handle double that and the temperature stays at 60° and doesn't climb.

The panels I use are made from aluminium (or aluminum if you're in the States ^^ ). They are 20 centimeters high so you could say I use the LookMumNoComputer Kosmo format. I bought 2 sheets of 1 meter long and 1.5mm thick and they are powder coated in gray/black. This powder coating is something I can really recommend because it's hard to scratch. If you just spray-paint your panels they will scratch very easily. You can write on the powder coating with a white acrylic pen. The one pen I bought had too wide a tip and I sharpened the tip with a razorblade but in the end it was un-useable. I ordered a pen online with a 0.7mm tip and that works far better. But if you laser-engrave your panels then you don't need all this anyway.

One IMPORTANT TIP I want to give you is the following: When you make your front panels for the modules set up a standard for their measurements right from the beginning. What I mean by that is decide on a fixed width for all of them. Choose for instance, 10CM for the bigger projects like the VCO's and 5CM for the smaller ones. Use those widths throughout the synth and don't do what I did and make them just the smallest size they can be. 
Here is why: You're going to build more panels/modules (eventually) than will fit in your synth cabinet and if they have a standardized width you can easily exchange them. For instance you might need more LFO's so you can take out a VCA and put in an LFO panel because they are the same width. I can not do that as easily because all my panels are designed with different widths.  
The thickness of the panels I use is 1.5mm That is thick enough and won't bend or flex when connecting patch cables. It also leaves enough room for the thread of the sockets and potmeters.

IF YOU BUILD THE MODULES ON THIS WEBSITE YOU WILL END UP WITH A MODULAR SYNTHESIZER THAT CAN RIVAL THE SOUND OF A BIG MOOG SYSTEM FOR JUST A FRACTION OF THE PRICE PLUS THE SATISFACTION OF HAVING BUILT IT YOURSELF!

Make sure the panels you are going to use are at least 1.5mm thick aluminium!! If they are any thinner than that they will bend or flex if you put a cable into a socket on the panel. So keep that in mind!
This cabinet is 1 meter and 11 millimeters wide, 38 centimeters high and deep. 
If you are building your own synthesizer I would be very curious to see some pictures so if you can link to that please leave the link in the comments. That'd be awesome!

POTMETER PIN NUMBERING:
To avoid confusion here is the way we number the pins of a potmeter:
Pin 1 is the Counter Clock Wise part (the left pin if you look at the potmeter with shaft facing you)
Pin 2 is the middle pin.
Pin 3 is the Clock Wise part (the righthand pin if you look at the potmeter with shaft facing you).

Here's an illustration of this:



WHAT DOES "U" AND "HP" MEAN IN SYNTHESIZER MEASUREMENTS?
When you're just getting into the modular synthesizer hobby you will hear/read terms like 'this module is so and so U high and so and so HP wide. What does that mean?
Well U stands for RU which means 'Rack Units'. It comes from the 19 Inch rack system. 1U equals one Rack Unit which is 1.75 Inch or 4.445 centimeters high. Eurorack modules for instance are practically always 3U high which is 5.05 Inches or 12.85 centimeters. There are also horizontal modules in Eurorack that are 1U high. Some rigs have an extra 1U row to accomodate them.
HP stands for Horizontal Pitch and is a unit for width. 1HP is 0.2 Inch or 5.08 millimeters. 
So the Kosmo panels I use are 4.5U high and a 10 centimeter wide panel would be 20HP.
Now there must be some rounding off of numbers going on in these standards because if I multiply 4.445 centimeters with 3 (for 3U) I get 13.335 not 12.85 centimeters. So I'm not sure what's going on there.  Maybe there are some differences between USA and UK standards. I don't know.

DUCKING and SIDECHAINING:
What does Ducking and Side Chaining mean? Ducking is a term for lowering the volume of some sound source so an other sound can better be heard over top. So the sound is ducking underneath the louder sound. This is a technique often used in percussion setups where, for instance, you need a kick drum to be heard over a bassline so the trigger signal that triggers the kick-drum also goes in an envelope generator which produces an inverted envelope that closes a VCA with the Bassline signal going through it. So as soon as the Kick drum is triggered, the Bassline is silenced somewhat so you can hear the Kick better. That technique with the VCA being triggered by the Kick drum is called side-chaining.

LEGATO: 
This is a term you hear a lot when talking about synthesizer or sequencer playing. Legato means that notes are played without the Envelope Generator being triggered by a Gate signal. The new notes you play will blend in with previous notes until the Envelope Generator's Release phase has died out and the synth falls silent or until a new Gate pulse is fired. There is usually a special setting on a synth or sequencer that makes this possible. Legato is usually achieved on a keyboard by not lifting the fingers from the keys completely before playing a new note. The sequencer in the Keystep by Arturia for instance can be programmed to play Legato. It will then play the notes you programmed in but without giving off Gate pulses for each note.

March 2020 the second stage:
Here are some pictures of the second stage of the synthesizer. This is a much simpler case and it is 20 by 20 by 100 Centimeters so it sticks out at the back a bit. This was necessary to accommodate the power buss system. I made some trunk locks on the sides so I can clamp the top section to the main synthesizer. This works just perfectly. I did have to solder these locks though because the locks themselves were connected to the main plate with the screwholes just by three flimsy bits of folded-over metal. So if you put any force on that they would bend and let loose over time. So I heated them with a blowtorch and soldered them from the inside. This worked really well because I used a bit of flux on the metal and this made the solder flow into all the little seems so it is very neatly soldered.





I made the width of the second stage too short by 1 centimeter so I had to use extra pieces of wood to connect the locks to.



EXTRA INFO: ABOUT THE STRIPBOARD I USE:
For everything I build and publish on this website I use standard stripboards of 24 strips high and 56 holes wide. The layouts I make are usually 24 by 55 so you have one hole extra room in case you make a mistake. You can order those stripboards from AliExpress for a very reasonable price. However, sometimes the pre-drilled holes are not quite in the middle of the copper strips which makes it difficult to solder components but this doesn't happen often and is the compromise you have to be willing to make. I have built my entire synthesizer with these 24x55 stripboards and all the layouts I publish use this size as a starting point. That's 6,5 by 14,5 Centimeter. Here's a link to where you can order them:



SOME EXTRA TIPS:

Here are some extra tips about general topics, not necessarily relevant to my projects but just things I want you to keep in mind.

ABOUT WIREBRIDGES:
All the projects here require wire bridges to be soldered in. Don't make those wire bridges out of electrically insulated multistrand wire, because it will get messy very fast on your stripboard if you use that. They take up a lot of room and create big solder blobs on the copper side. Avoid doing that if you can. Instead use single core copper wire, like transformer wire. Use sandpaper to clean off the insulating lacquer layer and reveal the bare copper and then you can easily solder them in place. I always neatly bend them to the right size too with a pair of neadle nose pliers. The picture below shows the stripboard for the Moog ladder filter (chapter 39) and imagine doing these wire bridges with normal electrical wire. It would be a mess.


MORE TIPS and INFO:
Measure the value of every component before you solder it in place. Most multimeters these days have transistor, resistance and capacitance modes so this should be no problem. It might save you a lot of time in trouble shooting later on.

Get an oscilloscope if you don't have one already! You are going to need one very soon if you go on building modules. Get a cheap one from eBay like a DSO138 for about 20 dollars. You can also look at the second hand market. Very good analog cathode ray oscilloscopes can be had on eBay for very little money. I myself decided to invest in a good digital scope and at that time the Rigol DS1054Z was just coming on the market and was praised to the hilt by Dave from the EEVblog on YouTube so I got one of those. I've never regretted that because it's a tremendous help to me and it has 4 channels.

Don't use lead free solder. I know, environment etc etc. The stuff is CRAP! Get the good old 40/60 or 37/63 Tin/Lead solder at a thickness of 0.6 or 0.5 mm. You can thank me later ;)

Why is output impedance important? Output impedance is a combination of the normal DC resistance and the AC resistance of any module that outputs audio signals. This is usually determined by a resistor in series with the output socket. Why is the value important? If you have a high output impedance and you use long cables, the capacitance of this cable combined with the output resistance (impedance) forms a lowpass filter that can cut off some of your high frequency audio. Don't ask me for details, you will have to Google that but this is one reason why output impedance is a thing. Normally we have a HIGH INPUT impedance because the inputs go into opamps which have an infinitely high resistance and we have a LOW OUTPUT impedance around the 1K Ohm mark among other reasons because of what I mentioned above. Now long cables are not usually used between modular synthesizer modules but they are used in amplifiers. I just want you to know about this, as part of you electronics knowledge.

Don't put those cheap Chinese Volt and Ampere meters into your power supply. I know these digital displays look cool and it's handy to know how much current your system is drawing but these meters introduce a shit-load of noise onto you powerrails. If you do want a measuring system in your powerrails, use analog meters with pointing needles instead. They look even cooler, especially when back-lit, and don't have internal circuitry that can introduce noise into your system.

Why does a filter have a Volt per Octave input??
There are two reasons why a filter has a V/Oct input. The first is that some filters can be used as sinewave oscillators if you set the resonance fully open so the filter begins to scream or gives of a loud whistle like tone. That tone can be made to follow the chromatic scale if you input a V/Oct signal and usually you also have to tune the filter using a trimmer somewhere in the circuit.
The second reason why filters have a V/Oct input is so that they open up more, become brighter, as you play higher up the keyboard. If the filter wouldn't do that then at a certain point your high notes wouldn't come through or they would be very much attenuated. So the higher the note you play, the higher the Keyboard CV voltage is and the more the filter opens up. 
Beware that there are situations where you don't want this behaviour, for instance if you're building a patch to make a Bass-Line. You want the filter to be low and dark so in that case you don't input keyboard CV voltage.

WHAT IS THIS BUSINESS ABOUT CERAMIC CAPACITORS AND WHY YOU MUST NOT USE THEM AS TIMING CAPS IN OSCILLATORS?
Ceramic capacitors come in a number of classes. Class 1 ceramic caps are also know as NPO or COG caps. These are the low value ones, upto about 1nF. They are relatively stable in keeping their value when the temperature changes. 
Class 2 SMD caps are usually higher values ones (1nF and up) and have an EIA code which gives you the lowest usable temperature, the higherst temp and the tolerance value in a "letter, number, letter" code. For instance X6R would mean -55°C to +105°C tolerance +/-15%.
Now the fact that these caps change their value with temperature makes them unsuitable for use in oscillators where you need stability to make sure they stay in tune.
But an other thing you probably didn't know is that these caps also change their value depending on how much voltage you put on them! Yes, I bet you didn't know that.

Now if you want to know more about this I advise you to watch the video below from the man that taught me almost all I know about electronics. Dave Jones from the EEVBlog:




MOSFETS:
I found a really cool video explaining how Mosfets work. Worth watching, he explains it very clearly with good examples. You'll learn about Gate capacitance, Inductions, Thresholds and lots more:


Vdd, Vss, Vee, Vcc, what does that all stand for?
These are abbreviations that you often see in schematics. 
Vdd (or VDD) = Voltage at Drain. This means Positive powersupply connection in MOSFET circuits.
Vss (or VSS) = Voltage at Source. This is Negative powersupply connection in MOSFET circuits.
Vee (or VEE) = Voltage at Emitter. Negative powersupply connection in BiPolar transistor circuits.
Vcc )or VCC) = Voltage at Collector. Positive powersupply connection in BiPolar transistor circuits.

MOTM
This is an abbreviation you hear a lot in the synth community. MOTM primarily refers to "Module of the Month," a line of premium, large-format modular synthesizers and kits originally introduced by Synthesis Technology.  Launched by engineer Paul Schreiber in 1998, MOTM pioneered a "no compromise" approach to DIY analog synthesis, utilizing high-quality, military-spec components. MOTM modules utilize the 5U height standard (8.75 inches or 22.2 CM tall) and are highly prized for their vintage aesthetic and professional-grade circuitry. 


Any questions or remarks? Put them in the comments below please. Comments containing links will be deleted!

Quote from Doepfer: "If I could change anything about Eurorack when I started it, it would be to use +/-15V instead of +/-12V." But we're stuck with 12V now. 

Wednesday, 20 November 2019

Synthesizer Build part-4: THE ENVELOPE GENERATOR or ADSR

This was the first Envelope Generator I built but I no longer use this design myself since I discovered the Digisound 80 ADSR and the Yusynth 7555 ADSR both of which are much better designs with the Digisound design also using the AS3310 chip (or the CEM3310). So my strong advise is not to build this particular design. "Why is it posted here then", I hear you ask. Well, because this website is an archive of the synthesizer I built and that includes all the lesser designs too. But you get ample warning if I advise against building. It's mostly the early projects that can be a bit problematic. The design below is simply the datasheet circuit of the AS3310 chip, which is not the best design by a long shot.

Original text of this article:
The Envelope Generator is generally better known as the ADSR which stands for Attack, Decay, Sustain and Release. These are the four phases a note goes through when you press a key on the keyboard. The attack is the speed of the initial rise of the note, once you press the key. Decay is the time it takes for the note to go from the peak attack level to the sustain level. Sustain is the level of the note as you keep the key pressed down. It is usually a bit less loud than the first instance of the note being played. Then we have Release and that is the amount of time it takes for the note to fade out once you let go of the key. So the envelope generator produces a signal that determines the volume of the note over time and this signal is being used by the Voltage Controlled Amplifier (VCA) which interprets it as an output level. In some Minimoog synths it is also called the Loudness Contour.

Now of course the envelope output is a control voltage so it doesn't mean that you need to use it for the above mentioned purpose. You can connect it to anything that can be controlled with a control voltage like the filter cut-off or the resonance or the pulse width of a squarewave. This opens up a miriad of options but let's not get ahead of ourselves here. If you're just starting out with synth building, you need the ADSR to open the VCA and the fancy stuff will come later.

I decided to build this ADSR using the chip series that I plan to use for the most important components of my DIY synth, the AS33xx series of chips. The AS3310 is the ADSR chip and it costs way less then its CEM counterpart. It's about €6,- 

So I looked up the datasheet and used the circuit that was presented there. I made the following stripboard layout for it. This layout is verified, I used it for my build. (All potmeters viewed from the front.):


(Last revised: 16-March-2020: Removed direct potmeter connection to 5Volt. Revised potmeter wiring. Trigger when not used shorted to Gate via internal switch in trigger input socket.)

Print only:



This design works very well and does the job it needs to do. It has a few little quircks though. The potmeters for instance. I used normal linear type potmeter and that works but it would be better to have reversed logarithmic or anti-logarithmic potmeters because the difference between 1 second and 10 seconds on the Release for instance is only a few degrees of turning the knob. But once you're used to this it's not really a problem. The Sustain potmeter is at its maximum at about 2/5th of its maximum throw. If you turn it further the Sustain level rises but the attack won't be able to reach it. So if you have Attack set to, for instance, one second, it will rise normally and after one second it will suddenly jump to the Sustain level. I've got some oscilloscope pictures below to illustrate this.
The external trigger input is normally shorted out through a switch in the Trigger input socket. So if there's no trigger cable attached, the trigger for the chip is provided by the Gate signal through the 3nF capacitor. The AS3310 needs a simultaneous gate and trigger signal to function. So if you plug a cable into the trigger input but you don't provide a trigger signal, the Attack parameter of the ADSR will not work! So this is not a malfunction, this is how it's supposed to work.

I added a few extra's to this design. First there is the option to output a signal that is twice the voltage of the normal output (10Vpp instead of 5Vpp). You can use this, together with a passive attenuator in the mixer panel, to drive a filter's resonance or other parameters of the synth. Then there's also an inverted signal output, to add to the options of driving parameters of the synth. This goes from 0V to -10V.
All this takes place at the opamp on the lower left of the circuitboard. You can use the old favourite TL072 for this or the TL082. Pins 1,2 and 3 together with the two 100K resistors form the 2x amplification. You can use other resistor values as long as you use 2 resistors of the same value. Then the input signal is split at the non inverting input and goes, via a 100K resistor to the other side of the chip where the opamp is set up as an inverting buffer. Both opamp outputs have their own output jack socket. The normal 5V ADSR output is a separate socket (of course).

And finally I added a manual trigger option, at first I added it so I could put a gate signal on the gate input for test purposes, but then I thought this would be handy to have anyway so I added the switch to the final panel. I decoupled the manual trigger signal from the gate input socket with a Schottky Diode so no voltage goes into the circuitry that delivers the normal gate signals so as not to damage it (although this is probably not necessary). I used a Schottky diode because their voltage drop is only 0.2 Volts so it doesn't detract too much from the usual 5 Volt gate signal.

Here's the E.G. mounted in my synth. You can see that I doubled the output sockets. There's 2 outputs for normal 5Vpp ADSR and 2 for either 10Vpp or Inverted -10Vpp I also installed a Gate output and a Trigger output. The Trigger output is connected to the Gate output via a 3nF capacitor and the Gate output is simply switched in parallel over the Gate input. I will however install a opamp buffer for the gate output in the near future.:



Here are some oscilloscope screenshots showing the function of each variable:
This is the normal envelope CV at 10Vpp.


Varying the Decay time:



The picture below shows the quirck I mentioned earlier where the Sustain level is set higher than the Attack can reach and so after the Attack cycle has finished the Decay is skipped all together (because there is no Decay if the Sustain level is higher than the Attack level) and the envelope jumps to the Sustain level. You can clearly hear the jump in volume in the audio. You can use this to your benefit though because it sorta has a percussive quality to it. Anyway, if you don't want this, just turn the Sustain down a bit. Problem solved. You can also limit the maximum resistor value of the Sustain potmeter by adding a resistor or trimmer to pin 3, but you'll have to experiment to find which value works best.



Lowering the Sustain level:



Increasing the Release time:



And finally switching between the inverse envelope (which was set to 0 to -5V in my ADSR but to 0 to -10V in the stripboard layout) and the 10Vpp envelope.



Okay, that's it for this one.
I hope you enjoyed this article and leave a comment please if you found this helpful! Much appreciated! Also, any questions? Put them in the comments or contact me on Facebook. I'm a member of the 'Synth DIY' Facebook group and the LMNC Discussions FBgroup and also the "Synth DIY for non engineers" Facebook Group all under my own name Eddy Bergman.
Stay tuned for more build reports and click 'follow' to be notified of new posts to this website.

Friday, 15 November 2019

Synthesizer Build part-3: TRIANGLE TO SINEWAVE CONVERTER

This article has been re-written at 11-11-2020 and updated in Dec. 2025.

After I had finished the VCO I wanted to add a Sinewave option to it. The first design I had posted here was a bit sketchy so I now present a new layout here. This layout has been made using the schematic of the Thomas Henry CEM3340 Deluxe VCO.
This circuit needs the Triangle input wave to be +/-5 Volt peak-to-peak. You can input a Trianglewave of 0V to +10Vpp but then the input must first go through the 1µF electrolytic capacitor to take away the offset voltage. The Triangle to Sinewave converter will not work properly if you input a 0/+10Vpp Trianglewave without first filtering out the DC offset voltage.

Here is the new stripboard layout. This converter offers a +/-5Vpp output and a 0/+10Vpp output. 


Here is the schematic drawing. I did not include any de-coupling capacitors but if you want to include them then just add two 100nF ceramic capacitors to the voltage rails as close to the chip as possible. One going from +15V to ground and the other from ground to -15V.
If you only need a bipolar output (+/-5V) then you can leave out the two opamps on the right and just tap the sinewave off at pin 1 of IC1-A as shown below, and leave out the 100K resistors and the opamps D and C.

Updated schematic 2025:


(There's also a KiCad version of this schematic further down this article.)
Here are two pictures from the oscilloscope. One without offset from the +/-5Vpp output and one with offset from the 0/+10Vpp output. If you look closely at the pictures you see that the scope is set to 2V per division and therefore that the amplitude of the sinewave is 8V. But now that I changed the feedback resistor Rf, that has changed to 10V (even a tiny bit over):



As you can see they are beautiful sinewaves and you can set the symmetry and distortion very accurately with the trimpots on the stripboard. 
It will be easy enough to mount this little stripboard on one of the M3 bolts used to mount the print of the 'Really Good VCO' and thus add a Sinewave output to that VCO. You can tap the Trianglewave straight from pin 10 of the AS3340 (or CEM3340) chip or from pin 12 of the TL074 quad opamp chip. I think that will be even easier. On those pins the Trianglewave is not yet given a +5V offset voltage so it is still +/-5Vpp and therefore doesn't need to go through the 1µF electrolytic capacitor on the layout of the Triangle- to Sinewave converter. 

Okay that's the new version of this article done. If you have any questions please put them in the comments below or on the EddyBergman Facebook Group page.

UPDATED VERSION OF THE SCHEMATIC Oct. 2025:
Here's the KICad version of this circuit which I used in the Eurorack PCB version of the 3340 VCO of project 18. It works like a charm and it has an extra 1K trimmer at the top of the transistor pair which lets you trim away any offset voltages. (I updated the stripboard layout with this feature too). It has a small range though so you need to make sure the 10K resistors are well matched to eachother. The transistors themselves must also be matched to get the best results. 


Here's a screenshot of the sinewave it produces. A really beautiful symmetric sinewave:


Btw, all the comments below upto August 2020 refer to the original Tri- to Sinewave converter article (which is now deleted) and not to this new updated one. So please disregard those comments as they are not relevant to this article.