Things With Strings

Last year, I snapped up a Vox Pathfinder 15 (non-reverb) combo amplifier from ebay. The seller was a charity shop, and the amplifier was listed as "untested", so I was taking a bit of a gamble, but figured I'd have a good chance of being able to fix it it it wasn't working, and cosmetically, it looked almost mint. When it arrived, I plugged it in, and as you guessed, it was not working. Barely any sound, and what sound there was was horribly distorted. One thin I did notice was the the power on LED was not aligned correctly with the hole in the back panel, so I did quickly took it apart, and realigned the potentiometers and LED indicator, but didn't have a look in detail at the circuit. I put it back together, but the problem remained. My guess was that the problem was the famous "cambridgitis" (just google that term and you'll find all about it). Basically, a problem where the optocoupler that controls the tremolo circuit fails. A common failure...

Having already modelled the preamp stage of the  Vox Pathfinder 10, 15 and 15R  respectively, and noted the similarities circuit-wise between the PF10 and the PF15, I discovered another much-loved practice amp with a similar preamp stage - the Marshall Lead 12. I should say, before going any further, that there are huge differences between the power amp stages. The PF10 uses a TDA2030A chip amp whereas the Lead 12 uses discrete components. The chip amp power stage on the PF10 shouldn't colour the tone much, but the discrete one in the Lead 12 most likely does contribute to the overall tone of the amplifier. A simulation of the preamp stage with the Treble, Bass and Mid controls at 50% and the Gain control at 0%-100% gave this Marshall Lead 12 Preamp Simulation Marshall Lead 12 preamp simulation, Gain 0%-100%, TMB @ 50% Vox Pathfinder 10 preamp simulation (overdrive engaged) Vov Pathfinder 10 preamp simulation, Gain 0%-100%, T+B @ 50% Comparing the two, ...

I've got a Vox Pathfinder 10 amplifier that I bought a couple of years ago, and I love it. It's a great little amplifier, but I don't care for the distortion channel, it just sounds harsh to my ears. There are quite simple modifications that can be done to this amplifier to remove the clipping LED's and increase the boost of the clean channel (check out Ben Craven's mods  for details). I'd been considering doing some of these, but whilst researching the subject, I discovered that the currently out of production (as at Nov '18)  Pathfinder 15/15R generally receives a lot more love than the Pathfinder 10. I found schematics for both the PF15 and the PF15R online, and saw that there were a lot of similarities between the PF10 and the PF15, particularly the earlier non-reverb PF15. The PF15R (with reverb) is quite a bit different after the first two opamp stages. I thought I'd model both the PF10 and PF15 pre-amp stages in LTspice  to see how they compar...

It's coming up to Christmas, and what to buy for your daughters boyfriend? Hmm, how about making him a pedal for his guitar. After much deliberation I decided on Tim Escobedo's Duende (MOSFET version) and drew up a super-compact perfboard layout... I went straight into soldering without breadboarding, as the circuit is pretty simple, and luckily it worked straight away. Here's a picture after adding all the components to the board. And here's a shot of the completed unit In use, I found that I needed the volume control at about 1 o'clock before I could hear any difference, and I was expecting to hear something a bit earlier on the dial, but once above 1 o-clock, there is plenty of volume to be had. Switching the device on does brighten the signal somewhat even at unity gain, and the device is very quiet in operation, No hiss, hum or buzz. I wish I'd had a girlfriend whose dad made pedals when I was his age!

I'm planning on building a bass drum module, and I will need a pulse generator to test it, so I knocked up a circuit based on a 555 timer in ASTABLE mode and put it in an Altoids™ tin. You can easily find a circuit for this just googling around. I drew one up in Kicad like this. As I wanted this to be a pulse generator for triggering a drum circuit, I wanted the frequency range to fall within a range that would be useful for music, which is calculated by a combination of R1, P1 and C1 in the circuit diagram above. There are plenty of on-line calculators that will help here. I used this one  http://www.ohmslawcalculator.com/555-astable-calculator  (P1=R2 in the calculation). I found that a combination of R1=3K3, P2=1M and C1 as 2µ2 gave the best range for my purposes (approximately 0.327Hz to 198.761Hz). I had a small piece of perfboard offcut that I tried to squeeze it onto, and here is the layout. For the output jack, I used a 3.5mm jack I salvaged from an old P...

After the success of Pete's Bazz Fuss  I figured it was my own turn. I've only ever used stripboard in the past, but wanted to have a go at building a circuit on perfboard instead, so wanted to choose something relatively simple for a first go. The circuit I chose was the EHX™Muff Fuzz™ which despite it's name os more of an overdrive than a fuzz type effect. I drew up my own perfboard layout as below, taking care to make sure the components could be connected easily when soldering. Soldering up was relatively painless, and I was quite please with the neatness. You can see from the pic below that it is a mirror image of the layout above. I used 2N5088's for the transistors and 1N60P for the diodes. The picture below shows the component side of the board prior to shortening the transistor legs.  It worked pretty much straight away to my relief, and sounded OK, so I boxed it up in a 1590A sized box (always a painful process) and here is the f...

It's been ages since I built an effects pedal, so to get my soldering chops back up to speed, I decided I'd build a simple fuzz pedal for my long-time friend Pete as a surprise present for his impending birthday. The circuit I chose was the Bazz Fuss (yes that's the correct name)  which you can read about here if you care to. I went for the basic v1.0 of the circuit and used a 2N3904 for the transistor, and boxed everything in a 1590A sized enclosure. Despite a number of stupid mistakes that I had to rectify, I got it all done and boxed within a day, complete with lo-fi graphics. Here's a gut-shot. gut-shot And here is the finished item. lo-fi graphic It sounded OK when I played it although it needed the volume knob maxxed out to get the best out of it. Maybe a higher gain transistor would have been better still. I'm pretty sure Pete will like it though.

I saw a piece of scrap offset in my bin that got me wondering about whether to try cutting some thin slices of wood and laminating curved sections. I tried to bend the piece of scrap around an 11" diameter MDF circle I'd made previously and I was quite surprised just how flexible the piece of scrap was. The scrap was varying in thickness, but one outer edge was about 0.15 inches. The photos show how easily it bent, and this was just dry, no steam or water involved. I might try laminating a few layers out of curiosity. scrap and 11"MDF disk test bend

After having all three layers done, I decided to try to get the middle layer (shallowest) to size using a router and template, then I would glue all three layers together and machine the outer two layers using the middle layer as a router follower bit against the middle layer. In short this turned out badly, as I broke the middle layer whilst trying to route the inside surface. Fortunately I was able to patch it up, and in the end the patch should be invisible, but I decided at that point that it was probably best to get the three layers glued together before I made an unrecoverable mess of things. glued and clamped I'm not 100% sure how I will shape it yet. Probably router plus jig for the outside and sanding drum for the inside. Part 2 is here

I'm still experimenting with things really, and decided to try a different approach for the next segment layer. This time I thought I'd make a octagon using 45 degree scarf joints. There are a number of reasons why. I'd had difficulty with the mitred angles on the previous layer, and using scarf joints means half the number of angles to using equal angles each side of the block. Half the angles means the cumulative error is going to be less if anything is off (in theory). The wood I have is a not really as deep as I would like, and doing it this way means initially I only need to get one side and one face squared. The other face can be fixed after gluing the segments. I don't have to worry about the length or width of each segment providing they are above the minimum requirement. On the downside, the glue-up is tricker as I found out later. So here is how it works out in sketchup 11 inch segmented rim using 45 degree scarf joints First off, I trued up t...

I was a bit nervous about 14 shoes being able to handle the tension, so I decided to re-model the bracket band with 18 shoes. This time I left a larder space between two of them to accommodate the neck, then evenly spaced the shoes around the remainder of the circumference. new bracket band model with 18 shoes For comparison, here is the old model. No idea why the colours are reversed, there's probably a logical explanation somewhere. old bracket band model with 14 evenly spaced shoes

Things are moving a little quicker now due to being able to square-up the stock faster due primarily to a jig I made for that purpose, hence I managed to get two layers done for the banjolele this weekend. I'd planned to work on the banjo rim, but avoiding nail holes, knots etc. left me with insufficient pieces for an 11" rim, but plenty for an 8" rim. Each layer is quite different in density and cloud, but I'm hoping it will hold together and look decent. The layers were 1" deep, as opposed to the 3/4" of the first one I made. The first one was quite yellow in colour, and I think I'll use that as the middle layer. It will probably end up looking like a cheese sandwich. Two layers glued and clamped I'm now quite looking forward to the next steps which will involve ensuring each layer is flat and then gluing them together. I'm not sure yet whether I'll sand the layers flat or try to use the router, I'll have to sleep on it. Part 3 ...

I'm only a beginner at Sketchup modelling, but have found it really helps to visualise what parts are going to look like, and to ensure everything is going to fit together correctly. I designed a bracket band which I intent to make from a 3/4" x 1/8" brass bar and 1/2" x 1/2" x 1/8" brass angle stock. My skills in Sketchup don't run to  rounding-off corners yet which I plan to do on the actual build, but the picture gives good idea of what I'm aiming for. Sketchup model of bracket band I've calculated that evenly spacing 14 shoes will just about allow for the neck to fit between two of them. Edit : I re-modelled this  Bracket band Sketchup model #2

When ripping thin slices of wood, it's a lot safer to have a zero-clearance insert in the table saw as it prevents the thin slices from getting sucked down inside the saw an possibly jamming the blade. Whilst out running around Christmas time, I spotted a piece of UPVC cladding that had obviously blown of the back of a truck and was littering the countryside. I thought at the time that it might be useful for making a table saw insert and made a mental note to go back for it which I did a few days later. Clean up the countryside and do a bit of recycling, how's that for a bit of good citizenship? It had suffered a bit of damage as it had bounced off the road, and it was pretty filthy, but it still had the protective plastic on it in places, and it was a fair sized chunk. UPVC cladding Here is the insert I made (white) by tracing around the original one (yellow). You can see the difference in the width of the slot that the saw blade comes through. UPVC zero-clearance...

Despite taking an unprecedented amount of care squaring the stock material and getting the angle on the mitre saw correct, the joints turned out pretty poor although it's difficult to see from the picture. banjo rim layer 1 However when driving home tonight I had a eureka moment and have gotten myself a solution. When it's dry, I will take a slice through a joint and insert a saw-blades thickness shim. Then I will let that dry and repeat the process for every joint. I may even use a darker colour to make it something of a feature. Sorted!

After a few practice attempts (aka false starts) at making rim segments last year, this weekend I had my first proper crack at it. The problems I'd run into previously were mainly due to starting off with non-squared stock material. This time I picked a ropey log from the pile of pallet wood I have (allegedly Mango, but could be anything) and squared it up. This particular piece was not the best, mainly consisting of knots, nail holes and worm holes and a little bit of wood holding everything together. I don't want to screw up on something that was too nice, but what started off as a fairly substantial piece only yielded enough wood to make one 12-segment layer for the banjolele. Anyway, once I'd squared up the stock, I needed to cut the segments at 15 degrees on the mitre saw. The calculation to get the correct angle to set the mitre saw for segmented is (360 / #segments) / 2 which is where the 15 degrees comes from in this case. First problem! My mitre saw had fix...

Before I get onto anything banjo related, I have to say how sad it was to hear of the death of David Bowie this morning. One of life's true originals, the world has lost one of it's most colourful characters. In part 1 I said I'd made a big mistake that would be revealed when I took the clamps off. When I cut the scarf joint, I somehow (goodness knows how) made a mis-calculation on where to start the cut. I think it's because I was trying to build in a margin for error, so didn't follow my drawing, and instead of a margin for error I just got an error. Pretty embarrassing for a guy with a degree in mathematics eh? Anyway, here is the unclamped piece, prior to clean-up. clamps removed Did you spot it? The headstock is too short (aaarrrggghhhh). Here is a close-up. the stubby headstock However, I realised after a few minutes of cursing, that it's actually not too bad at all, because correspondingly, the neck part is too long, and the headstoc...

Buoyed up by yesterdays scarf joint jig success and the cutting of the banjolele neck , I decided to start on the banjo neck. First off, I made an "improvement" to the scarf joint jig. A problem with the previous one was that because the part that was cut off was unclamped, there was a bit of tear out. So I adapted the jig in order that I could clamp the workpiece in two places. Like this... adapted scarf joint jig Yeah, I know what you are thinking, "he's going to cut his clamps in half". There is clearance though. clearance for the blade? However, I made two downright stupid errors! The first, and most stupid error was that I placed the clamp nearest the blade the wrong side of the cut. Duh! And the second was that I shouldn't have cut all the way through the fence, because I ended up having to place a wedge in it to prevent the clamp squeezing it shut. Anyway I did spot the second error before making the cut, but not the first one, ...

Continuing the banjolele neck I started yesterday (link here ) I couldn't wait 24 hours, so after 23 hours I took the clamps off. Actually, the reason was that I needed the clamps for my banjo neck, and I figured 24 hours was just an arbitrary number anyway and that if days were 23 hours instead of 24 it would have said 23. Anyway I digress. I hadn't got the clamping 100% right, because although the joint was good, the head was very slightly askew, so I cleaned it up on the table saw. I will sand the top level though before fixing the fingerboard. Here how it looks now. The scarf joint is very hard to see, it starts somewhere close to the fold on the cloth that the neck is laid on. I have to say that I'm fairly happy with it at this point. the neck after gluing the scarf joint

Delayed first by bird table, and then by the flu (not bird flu), I wanted to get back to some building. I took some measurements from one of my son's ukulele's and drew up some sketches for the banjolele in order to get the dimensions for the neck, and got to work on the build. First off, I cut off and cleaned up a piece of mahogany that I had. It's probably not real mahogany, just something in the same family. The piece I cut off was about 19" x 2.5" x 1.5" cleaned up neck blank Next, I made a jig for the table saw that would allow me to cut a scarf joint at a 15 degree angle for the headstock. I wanted this cut to be clean, so I put a decent Freud blade in the table saw. It actually turned out pretty good. 15 degree scarf joint jig Then it was time for the glue-up. This was a bit tricker than it looked, as it really wanted to move around when it was clamped, but I managed it in the end. neck and head clamped together Another 24 ...