CNC drilling machine update

Just a quick update after the postie dropped two rather heavy envelopes onto the doorstep of Nerd Towers this morning. The first parcel contained three 5v miniature stepper motors. These are really nice, and came from China in about 5 days. At £1.99 each with no extra p&p these could just be the bargain of the month!

The motors have lots of torque and because of the off-centre shaft, we suspect that they are really a much smaller motor inside the casing, with a number of gears to provide the torque. It's a cheap and easy way to increase the number of steps per full revolution, while keeping the power consumption/voltage/current requirements down so they can be driven through a ULN2003A darlington array (the kind we blew when driving stepper motors directly from a PIC last time!)

Also shown in the photo above are some drawer rails from eBay.
These are replacement parts for the kind of mechanisms commonly found in IKEA and MFI type flat-packed furniture. We got the 175mm lengths. About half of the runners run really smoothly and require little effort to push from one extreme to the other. About half have a little "sticky point" somewhere along their length. It's not an insurmountable problem, and our geared steppers should be more than enough to drive a platform mounted on these, but we're going to pick the four best/most freely running to use on our CNC drilling machine.
At £8 for ten off eBay, we've so far blown £14 of our £50 budget (we're going for the new build design to begin with, then see which parts we can swap out for salvage in a later version). Some laser cut acrylic, hot glue and bolts should see us with a working gantry in no time....

MIDI saxophone update

After another BuildBrighton meeting last night, a bit more work has been done on MIDI-fying our saxophone. This is now the third attempt and looks the most likely to get a working result! The first time, the wires were ran from the bell to the neck, leaving little or no room for the microcontrotroller circuit. The second time, the wires were run to the bell, so there was a large area we could fill with a circuit board, but the copper-tape-making-contact-with-the-body idea failed on number of pads.

This time, we're sure we've got it right. Only because, we couldn't have come up with a more difficult-to-implement solution! We're (slowly) removing each pad - yes, complete with linkages and springs - and fitting a pushbutton under each one.

In doing this, we're finding out quite a bit about the saxophone.
Namely that it's been badly dropped at some point during it's life, and would be unplayable as an acoustic instrument! There are four or five pads misaligned - and rings on the leather showing where the pad used to sit, and a shallower indent showing it's new resting position after the (presumed) accident.

So despite filling the body with hot glue and generally making a mess of things, at least we can be satisfied in knowing we're not ruining a perfectly good sax for this project - it was only destined for a junk shop somewhere, or an expensive repair job!

 (Saxophone fitted with five pushbuttons, mounted into acrylic disks. The push buttons are the slightly clicky tactile ones and the guard has been removed to allow the pads to open more fully). 

When operating the pads, it's interesting to feel the tactile clicky button through the button on the body of the sax. This gives a really nice, responsive feel.
However, some pads are "normally closed" and open when you press a key (or combination of keys) rather than the simpler push-to-close type. We found that sometimes, if you release a key gently, the spring on the pad isn't always strong enough to force the pad fully closed again. Here's an example of one such pad. An extra spring was added between the pad cover and the guard, to help push the pad into it's normally-closed state, and force the pushbutton to "click closed")

CNC drilling challenge

As part of the recent BuildBrighton MIDI workshop, we had to produce about twenty homebrew PCBs. The print and etching part was easy, but what took ages was drilling all the holes for the through-hole components.

There's no doubt that surface mount makes things easy, but when sharing designs and ideas, plenty of people are still firmly in the through-hole camp. So we try to reflect this when making boards for other people to use. Despite having a cnc machine at the 'space, and having access to a second one, the prospect of learning MACH3 and getting all the settings right to drill the holes with the cnc just left us cold. Also, getting everything out, set up, drill the boards, then dismantled and put back away again was going to be a big job (these cnc machines are big beasts and take up a lot of room, so we put them away when not in use!)

We felt it would be quicker for a few of us to just fit a tiny bit into a dremel and go for it by hand.

So this is what we did.

But drilling by hand was a real PITA too, so after this weekend, a few of the BuildBrighton members have set a challenge: To make as small and as cheap a cnc drilling machine as possible.

That description was a bit vague, so we've come up with some "rules":

• If buying all new components, not more than £50 on the entire build (just think about it, a cnc based device for under £50!)
• If using salvaged hardware (stepper motors from old printers, for example) not more than £20 total build cost
• As we're building a small machine, the footprint of the device, when put away, should not exceed A4 size (210x297mm)
• But it'd be useless if it could only drill tiny pcbs, so it has to be able to drill up to half eurocard sized boards (100x80mm)
• Any bought components should be accessible to everyone and you should reasonably expect to be able to purchase the same components from the same, or alternative suppliers 12 months from now (so you can't win a job lot of motors from some bloke off ebay and put cost of materials down as 50p)
• Overseas suppliers can be used to keep costs down, as their price per unit is often much less than UK suppliers.
• The device is to be platform independent - it can use any microcontroller and any software can be used to control the machine, including homemade software/drivers. 
• It has only to work on any one platform, not all
• For accuracy, the drilling machine should be able to drill a 1mm hole in a max-sized 2mm pad and leave a complete copper ring intact around the hole
• For building materials, assume a price for acrylic sheets as £1 per mm thickness, per A4 sheet
• Cheaper alternatives (such as mdf or laser ply) can be used instead of acrylic for the chassis if required
• Trivial components and the cost of pcb etching, running a laser cutter etc and consumables will not be used to calculate total build cost
• The price for delivery of all components bought online will be included in the final cost

More rules may be added as the build goes on, but only to clarify any points that may arise, not to act as a restrictive force to deter alternative ideas.

So there we have it. Another stupid challenge that probably won't get finished (or much beyond the planning stages) But it always nice to have an excuse to sit around with a brew and kick about a few ideas with like-minded people.

For our build, we're going for cheap AND small.
Most other people have concentrated on one or the other. We're more than willing to sacrifice accuracy for cost, or cost for size, or size for comedy value. As always, sporadic updates, in-between other projects, will get posted here as the build develops.....

Laser cutter back working

It feels like it's taken months (in truth, it probably has) to find a new home for the Nerd Club down in Brighton (well, Hove Actually) and to get the laser cutter back up and running.
But today we made some solid progress. With a PC donated by Chris at BuildBrighton and some cheap extractor fan ducting from Wickes, we finally got our laser cutter up and running in it's new home.

Of course, the first test cuts should be something simple like a few circles in each corner of an acrylic sheet, testing the laser across the entirety of the cutting bed. We decided against this and went straight into a commissioned piece - some personalised coat hangers for children's clothing.

Baby Elvis has some really cool threads to wear, thanks to his grandma. Now he has some cool coat hangers to keep his awesome new shirts in pristine condition (at least until next feeding time!)

We're just thrilled to see the laser cutter working after such a long time.
Hopefully this will be the first of many jobs to put through it over the coming weeks.

BuildBrighton MIDI Workshop

We had a great time over the weekend at the BuildBrighton MIDI workshop.
Not only did we cover the basics of Arduino programming, but there was even time to go through some electronics fundamentals - switches, resistors, potentiometers, voltage dividers and more!

The afternoon session was pretty awesome - once we'd de-mystified the MIDI protocol (note on, note off messages plus some excellent midi-thru examples from Jason, creating chord builders and arpeggiators) we got down to the serious business of making some noise!

It'd be great to say we had a jam and made some music, but we didn't really get much further than making cool noises, chord changes and whooshing arpeggios. But everyone on the course came up with their own unique ideas for implementing MIDI. At the start of the day, the most ambitious idea was a keyboard synth clone or a midi drumkit.

After the workshop, people were talking about light-to-sound modules, interactive art installations - even MIDI Flip-Flops. That's what the workshop was really always about!

(Jason in action during the BuildBrighton MIDI workshop, untangling a rat's nest of wires)

MIDI sax

It's late and it's Thursday, which can only mean it's BuildBrighton night again.
We've been busy working on our MIDI saxophone, trying to get it ready for next Saturday's MIDI workshop. Sadly, despite a promising start, it looks like we're going to miss the deadline. Here's why:

The original idea was to use copper tape on the insides of each pad and run a wire to a microcontroller. By detecting which pads were open and closed (copper tape touching the grounded body of the sax where the pad was closed would pull an input pin low, whereas an input pin linked to an open pad would be pulled high by a pull-up resistor) we could work out which note was being fingered, and play the appropriate note.

Sometimes it was necessary to add a little extra tape to the back of the pads, to stop the weight of the wire pulling the copper tape free. The copper tape is very sticky, but despite this, sometimes would start to peel away from the end where the wire met the centre of the pad.

With the first few pads, everything worked according to plan (with the exception that we had to file a little bit of lacquer away from the very rims of the sound holes, so that the copper tape could touch something metallic/conductive to make the circuit)

After completing the first few pads, we stopped and thought about where these inputs were going to. So far, we'd wired the entire instrument from the pads to the top of the instrument neck. This was a very small opening into which we needed to fit our microcontroller.

It was only when someone pointed out that we'd also have to use very small batteries that we realised our mistake! Instead of taking the wires to the top of the neck of the instrument, we'd have a lot more room (and could make accessing the battery compartment much easier) if we put the microcontroller onto a board and used it to "plug" or cover the opening of the larger bell part of the sax. D'oh.

It took just three minutes to undo about three days wiring work. But less than an hour to wire the first few pads up "properly".

As each wire went in, we closed the pad it was connected to, and tested for continuity between the end of the wire and the body of the instrument. This proved that the pad opening and closing could be used on an input pin of our microcontroller. All went well for the first six pads......

The seventh pad (on the left, below) was a bit more problematic.

This particular pad was not pressed directly by the player pressing a button, but was closed by a linkage from one of the other pads. As a result, it didn't quite have the same pressure upon it.
This was compounded by the face that this particular pad had a slight leak - it didn't sit exactly flush over the opening. When played as an acoustic instrument, this leak would cause the instrument to sound out-of-tune.

A more immediate problem for us is that because of where the tape sits inside the pad (sometimes fitting the copper tape to the underside of the pad is a tight squeeze and we can't always put it where we'd like) when this pad is pulled closed by the linkage, we're not getting the input signal going low.

It seems that, although relatively easy to implement, our copper tape idea needs a bit of a rethink......

MIDI bass frets

There's something not quite right with our laser printer.
It seems to be printing bitmaps at about 95% of full size.

We made up a couple of frets for our MIDI bass and allowed a generous border around each piece (so it can be cut/sanded/filed to an exact fit after being mounted onto the guitar neck)

The only thing is, that even allowing for this relatively large border around each piece, they are only just the right size to go on the fingerboard.

Here are some surface-mount resistors connected onto each section of fingerboard. These boards will be mounted face down onto the neck. The pads on the edges of each board will be connected by strips of copper tape, running under one, over the top of the  (plastic, non-conductive) fret between the two, and joining the pads on the next section of fingerboard. By doing this, we can create a resistor ladder for each string along the length of the guitar neck.

Arduino MIDI playground boards

In preparation for the BuildBrighton Midi Workshop this weekend, we've been busy etching and making up boards for the attendees to use with their Arduino microcontrollers.

These take the form of "shields" which sit on top of the main Arduino board.
Jason did a great job designing the circuit and PCB and Steve really made them look the business with his custom silkscreen/sticker job.

After etching and drilling, the stickers had to be carefully lined up. The easiest way to do this was to hold the board up against a bright light, and apply the sticker, taking care to get the pin holes shining through from the back into the correct places on the sticker front.

To make adding the components easier, we used a drawing pin to prick a small hole in the sticker at each point where a hole had been drilled on the PCB.  The end result was quite impressive!

After populating and soldering all the components onto the board(s) they're finally ready for testing before the big day. We wanted to make sure that we spent as much time as possible working around MIDI and actually creating something like an instrument, rather than lose half-a-day or more running a soldering workshop, so each lucky attendee will get a pre-made MIDI shield as part of their day's training.

Included on each board are
Opto-isolator (to conform with the MIDI specification)
4 x pushbuttons
8 x LEDs
1 x potentiometer (variable resistor)
1 x LDR (light dependent resistor)
1 x MIDI in DIN socket
1 x MIDI out DIN socket
Through headers to allow boards to be stacked on top of each other
1 x jumper to disconnect the LEDs to allow the digital output pins on the Arduino to be re-purposed

All in all, making the boards has been a lot of work, but we think they're worth it. They look really impressive when soldered up, and we're looking forward to seeing what ideas other people can come up with for them.

MIDI bass frets - exploded view

Here's a quick diagram showing how our MIDI bass fret system works.
We've already removed the fingerboard off our bass guitar, and plan to replace it entirely with home-made circuit boards (each section between two frets is a separate PCB).

The pcbs are fixed to the guitar neck, screwed down on top of some sections of rubber padding. This serves two purposes. The first thing it does is create a "pocket" for the surface-mount electronics on the PCB to comfortably fit without getting damaged. Secondly, it pushes back against the board mounted upon it.

This allows us to use copper table, placed over the fret (each fret is a piece of laser-cut acrylic) to create a single, continuous circuit up the entire length of the neck.
For each guitar string, the circuit is simply a resistor and LED, so each PCB has four resistors and four LEDs (one for each string)

BuildBrighton midi workshop

Over at BuildBrighton, we're working on putting together a make-your-own-MIDI-instrument workshop to teach people the core basics of MIDI messaging and how to send and receive MIDI commands. As part of the workshop, we'll be providing all attendees with an Arduino shield with some buttons, LEDs, volume/pitch bend pot and more.

As the workshop is just a few weeks away (on Saturday 23rd June) we've got some boards to make up!

Which means the nerd session or two is going to be taken up etching and drilling PCBs.

Using a mixture of press-n-peel blue and the cheap Chinese alternative, we've managed to toner transfer about 20 boards.

After a marathon etching session (about three hours!) here are no less than sixteen boards (the final three boards were still etching when this photo was taken)

All we need now are some spanky cool looking stickers (for the component side) from Robot Steve then we can drill and populate the boards next Thursday night.

MIDI bass fret setup

We're replacing the entire fingerboard for our MIDI bass, and replacing it with a number of PCB sections. We're basically creating a resistor ladder on the underside of the fingerboard, for each string on the guitar. The idea is that the resistor ladder creates a voltage divider and we take an analogue input from the resistor network into the PIC microcontroller

If all the strings are connected to ground, and all the frets on the fingerboard are connected at different points along the resistor ladder, we should be able to tell which fret the string is being pressed against, and therefore which note to play.

In the example above, if the (grounded) string was held on the second fret, the total resistance between the nut end of the board and the input pin would be quite high - so the voltage divider creates a high voltage on the input pin.

But if the player holds the (grounded) string on, say, the 18th fret, the 5v going through the resistor network goes through fewer resistors between the input pin and ground, causing the voltage on the input pin to drop. With clever use of resistor values, we should be able to create a look-up table in firmware such that we know instantly which string is being held against which fret. No more latency issues trying to sample and decode the frequencies that the strings are vibrating at!

MIDI bass instrument

As we haven't got everything quite to hand to finish off our multi-effect pedal circuit, and with Thursday being the regular BuildBrighton meet-up, we needed something else to work on. And, having access to a full equipped workshop with some really cool tools, it wasn't time to play about with resistors and microchips - we needed to do something BIG!

As we're working on some MIDI Arduino shields for an up-and-coming MIDI workshop, it made sense to work on a MIDI instrument. But exactly what?
We've already spent some time considering making a midi guitar - where the input from the strings against the frets is converted into a midi signal and fed into a synth. A few years we tried this, taking the audio output from the jack socket on the instrument, and analysing it to try to work out which note(s) were being played.

It turns out that this is very difficult.
To try to simplify things, we looked into making individual string pickups (so each string has it's own signal to analyse instead of trying to work out which chord is being played across all six strings!). This is exactly how the Roland's GK-3 "divided pickup" works. The signal from each string is sampled and an output value given to a different channel. The only problem with this approach is that there's a noticeable delay between plucking the string and the sound being played. It gets worse with lower notes and makes a bass guitar almost unplayable!

So we're back to using the contact of the strings against the frets as our signal, which removes the need to analyse the audio output. We started by:

Removing the traditional fingerboard from a bass guitar. This was quite easy (once we decided we'd never want to reuse the original fingerboard). Just plenty of heat from a hot gun, and some levering with a couple of screwdrivers.

It actually took three of us to do this! The neck was removed from the body and one person held it with one hand while hacking at it with a screwdriver with the other hand. The second person also levered away with a screwdriver, while the third waved the hot gun up and down the neck of the guitar. It took less than 10 minutes to remove the fingerboard!

(the scorch marks on the neck will be sanded out when we come to finish the instrument!)

With the fingerboard removed, it was time to design a replacement.
We're going to be making the replacement fingerboard from copper clad board, cut into sections, with an acrylic fret (3mm) between each one.

The first fret of the bass was measured as 46.5mm long.
Given that the length of a guitar string is exactly half one octave higher (i.e. the string length when the 12th fret is played is exactly half the string length of the open low E string) we came up with the following formula:

Length of next fret = length of previous fret / (2^ 1/12)
where 2^(1/12) is two to the power of one over twelve.

We made a cardboard template to try the idea out:

By the time we'd reached the end of the neck, we had 23 frets.
The original bass had only 22 so it looks like our original measurement on the first fret was a little out. However, it's still within a reasonable distance to recreate the original bass "feel" when playing.
Exact length isn't critical because we're never going to be playing the bass "acoustically" - our replacement frets will actually be switches, activated by the string being pressed against them. More details in the following post(s).....

Multi-effect pedal update

The Ferric Chloride yesterday evening, so we got on with etching our multi-effect pedal board, made with the (much cheaper) press-n-peel alternative. The result was pretty impressive. In short, the cheaper alternative works as well as the real thing for us. Which is just as well, as we've invested in about 50 sheets of the stuff!

A bit of drilling and filling and a bit later, we had a populated board!

Of course, the eagle-eyed will spot that there are still components missing off this board. Can you believe we can't find a 10K or 1K rated resistor in amongst all the junk in the nerd cupboard?! It's not like we're after some obscure rated component, just some common-or-garden, everyone-uses-them 10K resistors....

Here's the board soldered up (minus a few critical components) complete with "pass-through" switches. By setting these switches, you can have the input signal directed straight to the output jack (clean sound), have it pass through the fuzz circuit then straight to output, have the input go to the wah circuit only, or have the input go into fuzz then wah before hitting the output jack.

Still not quite ready for testing, but it's looking quite promising so far....

Components for onboard wah pedal

While waiting for our eBay Ferric Chloride to be delivered to test-etch some Press-n-Peel alternative a different parcel appeared in the post today - it's an order from Farnell for parts for our multi-effect pedal.
There's a fabulous selection of resistors, capacitors, buttons and switches:

Also included are some rather beefy 500mH inductors. They're for use in the wah part of the effect (just hope we've left enough room on the board as they're quite chunky!)

In fact, a quick trip to Maplin might be in order to see if our local Brighton store has any ferric chloride it (it probably won't - the Brighton store must be the worst stocked Maplins in the UK - unless, of course, they're all as hopeless as this!). I'd be great to get something at least working, if not finished by the end of the day!

Press-n-peel alternative

Press and peel is that great blue powdery paper used for printing circuits onto, to transfer onto copper clad board (either with a domestic iron or, in our case, a heavy-duty laminator).
It's easy to use, creates a sharp, clean image and is a tried-and-tested way of making circuit boards the world over. The only downside is that it's pretty expensive!

At nearly £20 for 5 sheets from Maplin it's nearly four quid a sheet! That's an expensive way to make your own PCB boards - per square inch, it works out nearly twice the cost of the actual copper clad board you're pressing the design onto!

Jason at BuildBrighton recently bought some alternative "toner transfer" paper from eBay. At £15 for 50 sheets (£20 with delivery), it's much cheaper than "regular" press-n-peel. Jason uses a household iron for his toner transfer PCBs and gets pretty good results (we have always found some smudging with an iron ,which is why we use the laminator approach)  But Matt has also used the same stuff and hasn't a good word to say about it!

There's only one thing to do - try it and see!
Jason gave us a couple of sheets to try out, so were' going to print and make up our multi-pedal circuit and see how it goes. We printed the PCB layout onto a carrier sheet, affixed the press-n-peel replacement and re-printed the circuit. Here's the result:

The first thing we noticed is that the image is very delicate to the touch. In places, the black toner is already flaking away. While this is probably good news for the transfer part, it's not great for making a stable image on the paper. This was printed using the usual settings - a Xerox Phaser 7400 printer with commercial press settings, full black with the fuser set to "thick card stock" (we use 360gsm card as the carrier sheet).
(the photo above was taken as the image came straight out of the printer and before we poked and prodded it to see what the effect would be!)

The image was transferred onto the copper board using three passes of our laminator, as we usually do with Press-n-Peel:

The final image was quite stable - rubbing a finger over the copper removed no further flakes of toner - the image looked pretty much the same as it did on the paper before transferring. The flaked parts of the solid plan can easily be filled in with a Sharpie (permanent marker) before etching, but the important thing is that all the traces are solid and there were no immediately obvious trace breaks or blurring which would result in a badly etched board.

We're waiting for our ferric chloride to be delivered following a last-minute eBay order, so we'll have to etch the board as soon as it arrives, to complete the comparison between this "cheap alternative". But to date, it looks like it could be a suitable alternative.

Press n Peel blue = £18 for 5 x A4 sheets from Maplin = £3.60 per sheet
Press n Peel blue from eBay = £11 for 5 sheets (including delivery) = £2.20 per sheet
Alternative = £19 for 50 sheets (including delivery) = 38p per sheet

At almost a tenth of the cost of Press-n-Peel blue from Maplin and nearly a sixth of the price from the cheapest alternative supplier on eBay, we're really hoping that the final etch is as good as the transferred image! It's all well and good getting a transfer as good as press-n-peel blue, but it's the etch resistance that we need to be sure of.

Hopefully we'll have an answer tomorrow.....

MIDI saxophone

This rather tongue-in-cheek article (http://www.mindworkshop.com/saxophone.html) explains how I feel about learning saxophone. I bought one a few years ago, even took a few lessons, learned a few songs, but then put it back in it's case, where it's stayed ever since.

Why? I guess I'm just too considerate!
I didn't like the idea of my neighbours having to listen to the squeaks and farts that came out of the shiny bendy bit day and night, and the cost of going to someone else's house to learn and practice become prohibitive quite quickly. So I figured that one day I'd build a practice sax, that would let me learn the fingerings and notes without having to blow the thing too loudly.

That was four or five years ago.
My lovely saxophone was too nice (and too expensive) to hack, so every now and again I'd check eBay for a cheap instrument. Well, this morning, all that waiting finally paid off

Here's my new sax, ready to be fitted with a load of wires to be converted into a midi instrument. My first idea was to run some CAT5 cable through the body, then separate the wires afterwards. This very quickly proved too difficult

So I separated the cable into individual strands and placed a small weight (capacitor) at one end of each wire.

I then pushed the capacitor through an open, erm, valve? pad? hole? and wiggled it round, while feeding the wire into the body of the sax.

Eventually it appeared at the neck end of the instrument (I started feeding the wire into the larger holes in the curved bit at the bottom of the bell and worked up the instrument)

Sometimes - especially once a few wires were in place - the cap got caught up with other wires or fell into the hole of another note. If it couldn't be shaken free, a bent bit of solid-core wire was easily pushed down the next to hook it free and retrieve the other end of the wire.

After about an hour of wire-wiggling, I managed to get twelve wires from the lower 12 holes, through the instrument body and secured out of the top of the neck. Once all 24 are in place (the body has 23 holes and a 24th on the crook of the mouthpiece) then can be soldered to a PCB and hooked up to a PIC microcontroller for the MIDI fun to begin!

Multi-effect pedal

We had a great time at BuildBrighton with our working fuzzface guitar pedal. New member Patrick even brought along a VOX wah pedal to get some Voodoo Chile type riffs going!

Which got us thinking about embedding our effect pedal inside a guitar, and how to simplify the circuit. So we came up with this idea - it's an effect pedal combining a fuzz face and a wah pedal. Either or both can be by-passed (so you can have clean, just fuzz, just wah, or fuzz going into wah).

In a future version, we're planning replacing the potentiometer on the wah (labelled VR3) with a digital pot. This will allow us to hook the wah effect up to a PIC micro-controller and use a 3-axis accelerometer to change the wah sound. Tip the headstock towards the ground to simulate rocking the pedal back, then move it upwards to rock the pedal forward.

The introduction of an accelerometer on the guitar opens up all kinds of possibilities (imagine hooking it up to the volume control: tip your guitar skyward, a la Slash, for a volume boost while solo-ing!) Any way, here's the first version of our multi-pedal:

Multi Pedal schematic

Multi Pedal PCB

Parts list:


C1 2.2uf  capacitor
C2 0.01uF capacitor
C3 22uF capacitor
C4 0.01uF capacitor
C5 0.22uF capacitor
C6 4.7uF capacitor
C7 0.01uF capacitor
C8 0.22uF capacitor
L1 500mH inductor
Q1 BC108 NPN transistor
Q2 BC108 NPN transistor
Q3 BC108 NPN transistor
Q4 BC108 NPN transistor
R1 33K resistor
R2 330 resistor
R3 8K2 resistor
R4 100K resistor
R5 68K resistor
R6 1.5K resistor
R7 470 resistor
R8 470K resistor
R9 22K resistor
R10 33K resistor
R11 82K resistor
R12 470K resistor
R13 10K resistor
R14 1K resistor
SW1 = DPDT switch
SW2 = DPDT switch
VR1 = 1KB potentiometer
VR2 = 1KB potentiometer
VR3 = 100K potentiometer