My Amp Building Journey

Note

This is a blog that I originally published over at GuitarMasterClass.net where I was an instructor from about 2006 - 2009.

Aug 14, 2007

I have spent a month or two now trying to find out about the finer points of Guitar Amplifier knowledge and terminology. Yes, I am on a quest for tube tone - I have been convinced that there is more to life than than Line 6 and I am setting out to find it.

I was pondering Carvin and Bogner, Marshall and Vox when I saw a post on the guitarmasterclass.net forum - a member put up a picture of a tube Amp he had made … then it hit me, why couldn’t I do the same?

So with that goal in mind, I am going to build an amp, maybe several, until I get the sound I want. I have a lot of experience in electronics (built all sorts of gadgets as a kid) but I have absolutely no experience with tubes and tube circuits, although I do understand the physics of them.

Research

I have been looking around the net and have found a lot of websites that sell kits, but I wanted a bit more than that - I wanted to understand how it all worked and maybe tweak things a little to get the sound I have in my head. Adam pointed me to 18watt.com - they have a number of projects on the go, their first design was an 18 watt Marshall like tube amp. I had a look around but found the website a little hard to navigate, didn’t really want a Marshall sound and was generally uninspired by it all. After some more searching, I happened upon AX84.com. A really cool website with a very active community. Over the last 10 years they have designed a prototype tube amp (the P1), along with many variants, and are now on a grander work of providing a library of reference designs consisting of Pre and Power amps you can mix and match. I liked the site and decided to build one of their simple designs - the simplest have just a couple of tubes and a few tens of components, but building a tube circuit can be tricky and I wanted an easy place to start.

Note

As I re-athor this in 2025 I sadly note that AX84.com has ceased to exist, although there is an archive for it at the above link.

The Amp For Me

Part of what I wanted was to explore the different sounds out there. I have zero tube amp experience. None, NADA. I don’t know Marshall Brown from HiGain from Vintage from a Vox - so this is a huge learning curve for me. I have had some good advice people on the GMC forum and know that:

  1. Tube Amps are loud
  2. They have to be played fairly hard to get all of the tube benefits.

A 35 watt amp with an efficient speaker at 10 feet can rival a jet plane landing for decibels, so something that you can play high enough to get the right sound yet record and practice with needs to be pretty small indeed. The AX84 P1 is a 5 watt design based on a 12ax7 double triode for the preamp and an EL84 for the power amp stage. I looked at it initially, but although it was simple, it was basically a crunch type of amp, and I am looking for something higher gain, although not monstrously high.

The P1 has a couple of variants, the Extreme, and the High Octane. High Octane is a more complex build but has an EL34 or 6V6 power tube, a 3 stage preamp with cathode follower which makes it a high gain amp. The Extreme still only has 2 preamp stages like the P1 but does benefit from an EL34 or 6V6 power tube. This was appealing as I like the voicing of an EL34 based amp but would also like to experiment with 6V6 - normally hard to do as to exchange tubes you need to re-bias which needs knowledge of the workings of the circuits. The Extreme also keeps the simplicity of the P1, but can be upgraded to be more like the Hi Octane if I need more gain.

So, my choice was made, and I ordered the parts this weekend. I got a kit from Doberman Amps and a pre-drilled chassis, since my metalwork is dire. I also upgraded to a better output transformer and a variable bias option so I can easily swap power tubes. It has cost me around $300 so far, and counting …

The Speaker

What little research I did on the internet seemed to point to the Celestion Alnico Blue speaker being the holy grail - they were used extensively in Vox amps for a while where they gained their reputation. Not cheap at $250 a pop, but I am doing this to get tone, not necessarily to save money so it seemed worth it.

Other Bits and Pieces

I’ll need something to put all of this in, and I am looking into building a combo cabinet - unfortunately my woodworking skills also totally suck, but maybe this is a chance to learn some other stuff and do it right, I’ll think about that some more. I could get a nice tweed style combo cab ready made for about $300 - could probably build it for 75 - 100. Still thinking about this.

Aug 17, 2007

Well my speaker has arrived - looks like it survived the trip intact, here it is

Lovely and blue, hope it sounds as good as it looks …

I have been talking to a guy on AX84 that might make me a custom case for the amp. I fancy a tweed style because it looks distinguished - something a little like this maybe:

I might have a go at building it myself depending on price, but my woodworking skills are as I said, dire. I have been reading up on cabinet construction though and have almost convinced myself that I could do it (a decision I have regretted in the past!)

Tweed is expensive though, I figured about $100 to cover a mod sized combo. The other alternative is Tolex but that is a little utilitarian for my taste. I can buy a ready made tweed covered cabinet form mojotonefor around $300 plus any mods I would need to fit my specific construction. Still mulling this over really.

I have also researched a couple of sites that will make face plates to a design so I can get some cool labeling for the controls.

Aug 27, 2007

Since I last posted I have positively ruled out the prospect of building the cabinet myself … I had almost convinced myself that I could do it and that previous disasters had just been bad luck. However, 5 hours in the garage with a lump of wood, Black & Decker Workmate, circular saw and a ruler have now convinced me otherwise. It took me several hours to cut 4 pieces of wood - this was just an experiment. They were supposed to be the equal four sides of a cabinet. Long story short, equal they were not. Cabinet would have been wonky. Forget woodworking, my talents lie in other areas.

I’m still liking the tweed style case, but I can buy one for $300 - $400. Expensive but it should look good.

In other news, the kit arrived this week:

Lots of lovely parts to fit together. The pre-drilled chassis came with it and today I lacquered it.

The timing is good - I have 5 days off work over next weekend. I’ll be away for a couple of those but that leaves me some time to start hooking this beauty together and realize my path to sonic Nirvana.

In addition to the kit I have ordered some PTFE coated hookup wire, and a decent Fluke Multimeter for testing once it is made. I also ordered a Weber MASS power soak for low volume recording and practice so I should be all set when I have put it together.

Sep 3, 2007

Well after a busy week and a trip up to Massachusetts I finally have some time to put into this project!

I am still waiting for my wire to be delivered - it got caught up in the vacation and should arrive tomorrow. Not too bad as I am not returning to work until Wednesday. With a little luck it will arrive early in the day and I can actually start soldering some stuff together!

My Fluke Multimeter arrived, and now I feel like a real professional - I have always wanted an excuse to get one of these. I plumped for the Fluke 115, by no means top of the range, but far from entry level too. I have also ordered a few valves and sockets and some other hardware from The Tube Depot - a very impressive mail order outfit. They have a great selection, and shipping took just a couple of days. I ordered some EL34s, some 6L6s, a couple of spare 12ax7s and some EL84s just for kicks, along with a number of tube sockets. (The latter is because various of the mods I have in mind for my amp involve additional preamp tubes - the best time to wire in preamp tube heater circuits is at the beginning, its a lot harder later on, so I am building in a spare socket and wiring its heater circuits from the outset).

In the meantime, I have mounted all of the chassis components:

  • Transformers
  • Input/Output Sockets
  • Tube Sockets
  • Switches
  • Fuse holder
  • Potentiometers
  • Power Socket

Here are a few pics to be going on with

Top:

Front:

Inside:

Its starting to look almost like a real amp now

Next installment (hopefully tomorrow) is the circuit board, followed by the chassis wiring. This is pretty involved with lots of carefully twisted and routed wires so I won’t rush it.

In other news, I have emailed Mojotone about one of their combo cabs - hopefully I will be able to fit everything into it but I need some measurements from them first.

And finally, I bought a tube tester off eBay - a big old thing built in about 1960 but it still works - amazing!

That’s it for now, all things being equal I hope I’ll have an update tomorrow!

Sep 4th, 2007

Cool - my PTFE coated wire arrived early this afternoon allowing me to swing into action!

I don’t have a permanent electronics lab yet - need to move before I can get that, I set up a table in the garage, looks pretty technical if I do say so myself!

In the picture above, you see the tagboard just before I started to solder all of the board mounted components on it.

About 2 hours of hard work later, I ended up with this:

Although I had done a lot of soldering in the past, I had never used an eyelet board like this. I would have preferred a turret board actually, but this is what came with the kit of parts and turret boards are hard to fabricate without the right tools, chief among those being a drill press which I do not posses. It all went well, including soldering of the underside wires that you can’t see in the picture. Next step will be to wire he board up to the chassis components and a whole heck of a lot of other wiring. I am planning to make a start on that next weekend, for now I am pleased with progress.

In other news, I have been corresponding with Mojo Sound about a case for this beast - I have narrowed down what I want and I am waiting for a quote. I am now expecting it to be around $400 because we are into custom cases - their off the peg case wouldn’t quite do it.

Sep 5th, 2007

Having thought about it a little more, I’m not 100% happy with my efforts above. The joints were fine and electrically sound I am sure, but I did not like the eyelet board much. It will only get worse as I start attaching wires - some of those holes are already running out of space. Add to that I am using 18awg wire - only necessary for the heater circuits but I figured thicker wire would be better so decided to use it throughout.

After some thought this morning and a little research, I decided to redo this part of the project as a turret board. In terms of time this is a setback - a waste of 3 hours yesterday, and an additional wait for parts and another couple of hours to make the board and another hour or so to re-solder it. But I am not in a rush, and I am treating this as a learning process. I want people to be impressed with my amp, and I am just not happy with the look of the above board, it looks amateurish. In financial terms this isn’t a big deal. I’ll have to re-order all the parts that went on the board, but they are fairly cheap - resistors are about 5c, even the larger caps will only be a couple of dollars each. Probably the total for new component will come to $15, so no big deal. Turret board and turrets are pretty cheap, the only real expense will be a drill press - you can get those for $50 and up, which was cheaper than I thought, plus I will be able to use the press to drill my own chassis for my next project whatever that is.

With a little luck, if I order the parts today I will have them for the weekend. If not, there is some chassis wiring I can get started on even without the board.

In other news, Mojo have been very helpful in working up a cab spec for me, and the final cost is $352.50 which was cheaper than I thought, although shipping will probably add a fair amount to that.

Sep 6th, 2007

Today I ordered a drill press. Its got a laser so it must be good. I sent an order to Mouser Electronics for new board components - the big caps were about $10 each - ouch! So I revise my estimate to about $50 of components and some more for the turrets and board. It will probably be after the weekend before I get all of this together, so my weekend task will be to wire up the chassis, I can use my existing board as a guide, I just won’t hook it up. Once all the wiring is done, it shouldn’t take long to hook all the wires up to the new board when I have built it.

Sep 8th, 2007

A big push today - I spent the day wiring up the chassis, ready to drop the new board in when I have it done. Still waiting parts for that, but the wiring went well. All told it took me about 6 hours to do this - partly because I was getting used to making the twisted looms, but also because there is a lot of nitpicking, and I wanted to do a good job of routing the wires. It doesn’t look a lot for 6 hours work, and looks messy because there are a lot of loose wires waiting to be soldered to the board, but I’m pretty pleased with it. The hardest part was the heater filaments (green and grey twisted wires) - I didn’t get those perfect but they should work fine.

Next step will be to make the new board - first I have to drill it and fix the turrets in, then wire all the components in place. It will be probably 3 weeks before I get around to that as I have a vacation planned and I’m still waiting for parts and the drill press, but another couple of weekends and I should be all done.

My plan is to build the basic amp then add some mods - on my list for thinking about are:

  1. Switchable cascode boost stage - should give me something approaching a high gain sound
  2. Swithchable triode/pentode - I’ll try this but people have reported that it doesn’t sound too good on this amp
  3. Switchable Bypass capacitor - Fat, normal, bright
  4. Switchable EL34/EL84 - a tough one this as I have to figure out a way of dropping the anode voltage by 100v
  5. Effects loop - simple serial unbuffered would be easy, or maybe I’ll use an additional 12ax7 and buffer it
  6. Conjunctive Filter - an additional capacitor and resistor across the output transformer primary that smooths out the frequency response a little

All of these are moving away from the safe world of tried and tested schematics though, so the most likely thing that will happen is that I’ll blow something up!

Sep 24th, 2007

Ok, back from my vacation. While I was away, I took delivery of an oscilloscope and signal generator. Neither are essential for building an amplifier, but they can be useful for scoping input vs output, measuring gain and also for setting and checking tube biases. Besides, I wanted an excuse to get another Oscilloscope! The Scope was bought from a nice guy on AX84.com, and I bought the signal generator from Ebay.

Its fairly basic - only 20Mhz, which is great for audio but not up to digital and radio applications, but the price was right and it will do everything I need. Its used, but in good working order.

The other big news is that the cab arrived from Mojotone. It looks really cool, and seems to fit my chassis pretty well - I had a long email discussion with Mojo about dimensions and they helped me to specify exactly what I want. The only slight issue is going to be travel of the knobs I think, but I might get away with it.

The next step was to mount the speaker …

Looks good in there :)

Now I need to get through this next week, and I have some more vacation time around next weekend. I am going to see Steve Vai in Atlantic City, but there will be a lot of other time to work on my amp. I have to build the turret board, solder in the components and hook it up to the chassis wires - then I will be into testing. I am hoping to get a lot of that done around next weekend.

Oct 6, 2007

Busy few days … Steve Vai was awesome and the trip to Atlantic city was a lot of fun. Work reared its ugly head however and I had to cancel most of my remaining vacation. This weekend however I made some progress. First, I assembled the drill press - not as trivial a task as you might imagine :) It was pretty cheap but does the job really well. Surprisingly, the laser is a lot more than a gimmick as it lets you pretty accurately place the thing you are drilling. I did a test drilling of some fibreglass board to see how it would work out. Went pretty well, but my holes were not in straight lines - the trick for that is to clamp a piece of wood to the press as a straight edge - I’ll do that for the real board. Here is a pic of the assembled drill press, drilled board and the “X marks the spot” laser sighting.

Next step would be to place the turrets using the turret tool. Unfortunately however, although I have the tool there is no way to secure it in the drill press. I have ordered a drill press vise which should arrive in time for next weekend and I can try and fit a few turrets to make sure that is all ok. If that works out I’ll drill a new board with the straight line technique and fit the turrets.

October 21, 2007

This afternoon I completed the turret board - hurrah! I’m now in a position to rebuild the board to the higher standard I crave. With the correct tools, making the turret board turned out to be pretty easy, I would say an hour from start to finish. The first step was to accurately place the plan on the board as a drilling guide.

The drilling plan wouldn’t actually fit on one piece of letter sized paper so I printed it in 2 parts and taped them together, then taped them to the board. Next step was to set up the drill press with a piece of wood for a guide to aid drilling in straight lines.

This worked pretty well in conjunction with the laser, and allowed me to quickly drill the board.

Much better than the test board, all holes are pretty much lined up. Note the slightly larger holes - these are screw holes to fix the board to the chassis.

As is always the case, my efforts to order everything I need online are doomed to failure. The drill press vice arrived but it come with no way to fix it to the press. A trip to home depot later, and I had a set of bolts that didn’t fit the press either - duh :( However, knowing my predilection to fail embarrassingly to get the right sized pieces, I also bought some washers, and they ended up being a far better way to secure the turret tool in the drill press. With this in place it was an easy matter to insert the turrets and press them into place.

Finally, after half an hour of stamping I have a finished board.

I called it a day there - now I can go back to soldering the board, and wiring it to the chassis, another day!

October 28th, 2007

Building upon my successful board creation of last weekend, I have now completed the soldering of the main board - which puts me back where I was many, many weeks ago. This whole exercise is becoming somewhat episodic as work and home commitments phase in and out, preventing me from the rapid progress I had originally envisaged. Be that as it may however, I am still just as motivated to finish this as I was in the beginning, I now see that I am in this for the long haul. I will be moving house in 3 weeks or so, and everything will become frantic around here, so I doubt I will find a lot of time to work on this until after that madness is over, however, when in the new house I will have a lot more space for working and actually prefer the idea of initial switch on and testing in my well appointed new garage with permanent bench bristling with test equipment, rather than the folding table I have now.

In any case, after my construction of today, my board looks like this:

I am a lot happier with this turret board based effort than my previous version, it looks better and I believe the construction is superior - it is much easier to make decent solder joints this way, so it was well worth the effort and the delay. As is always the case, when I ordered the new components I missed one out - a 330 ohm, 5 watt resistor, which was unaccountably absent - I’ll have to order that in the next week or so. Actually, this is probably a blessing in disguise, because in the first version of the board I used the wrong value, so I am glad I caught it this time.
After completion, it was a simple matter to drop it into the chassis.

Although I was tempted to continue and start hooking the wires up, I decided against it. To avoid destroying components, it is advisable to commission an amp in stages, checking the heater circuits first, then hooking up the high voltage rails in a sensible sequence, so I will save that activity for when I have the time and equipment setup for it.

December 26th, 2007

OK, its been a while - the move went smoothly, but was a lot of work. After I dealt with the essentials, such as somewhere to cook, and sleep, then the nice to haves (somewhere to watch TV, an office to work in) I was free to start work on the garage, or electronics lab as I like to call it now :)

The garage itself is actually a single garage, but is on the roomy side - with our car in there I still have a lot of room for a bench. In fact there was a bench there already, and although I will ultimately put a nicer one in, it definitely does the job for now. The only thing I really needed was a work light as the only sources of lighting in there were incandescent bulbs behind me, so they cast a lot of shadows. A quick trip to Home Depot solved that - I bought a large fluorescent lamp and fixed it right over the bench, and I now have a huge amount of illumination - we’re talking Sahara Desert at midday here! I spent a few hours clearing out all the junk that had accumulated in the Garage during the moving in process, and unpacked my scope, signal generator and other tools.

Next problem was that there was no power to the bench … I solved that by running an extension cord to a power strip. Temporary, but it will work for now. I have in mind a detailed garage project for the future which will involve a better bench, some nice storage shelves and a permanently wired electrical outlet. For now however, I have all I need to get back to the important work of crafting my Amplifier. Now all setup, the Electronics Lab looks like this:

So, now I had a lab I could pickup construction where I left off - you will remember I had just dropped the board into the chassis. The next step is to wire it up, which I did. It took around an hour and was fairly routine - I wired in the tonestack and the leads from the board to the tube sockets as per the schematic - a great way to ease my way back into the project.

When powering up a tube amp for the first time, it is advisable to do it in stages, so a number of leads weren’t hooked up - I will do that as part of the big switch on in the next installment! So now it looks like this:

I didn’t attach:

  • B+ Lines
  • Heater Lines
  • Output Transformer Secondaries

If all goes well, the next update will involve getting some actual sounds out of this beast - if all goes badly it will be clouds of smoke and some fizzing noises …

January 1st, 2008

My previous words were prophetic - as it happened I got both sound and fizzing noises - let me explain …

After a few final solder joints, and a quick check over of the layout and components, I launched into Paul Ruby’s excellent Amp Power on Procedure. Paul is a man who has built many amps, and came up with this checklist and step by step process to powering up a tube amp for the first time without destroying all of your expensive parts. I took it by the book.

You have seen a few of these pics before, but for the record, the final working version looks like this - all leads soldered, everything in place and working. I have seen neater builds bit on the whole I am satisfied.

My first step was to build a bleeder lead - a crocodile lead cut apart with a 100 ohm resistor soldered in between the 2 ends. The idea is that the bleeder resistor drains the filter caps in between tests so when you are moving probes and clips around you don’t get a 400v belt! The P1 Extreme design has built in bleeders, but this stuff is dangerous and resistors have been known to go open circuit. As a further further precaution, I connected a spare meter across one of the filter caps to make sure the voltage dropped to safe levels after each test.

With the bleeder lead in place, I started down the list:

  • Various checks for shorts
  • Correct voltage from power transformer heater circuits
  • Correct B+ voltage when open circuit
  • Correct B+ voltages when connected to the board

This all checked out, so I finally put the tubes in and switched on. All seemed fine, so I plugged in my guitar and was rewarded with some glorious tube tone! Yep, it worked perfectly first time, I am pretty proud of that. No unexpected hums, buzzes or oscillations, all the controls seemed to work as expected. So of course, I dimed it … I really enjoyed the pretty white light at the base of the EL34 until it fizzed and the amp went dead. Five minutes of playing and my amp was dead.

Of course, I had a spare tube, but I was reluctant to just drop it in, in case there was something intrinsically wrong. I posted on the AX84 website, worried that my failure to set the bias had resulted in the tube failing, but was reassured to hear that if that had been the case I would have seen a red glow on the plate of the tube, not a white glow from the base. The working assumption was then that the initial tube was a little flakey. I broke out a replacement and powered up again, this time I set the bias to a reasonable value, by measuring the voltage across my bias test points (the red and black terminals you see) and adjusting the potentiometer to get it to a reasonable value. The correct bias will vary with every tube you use, but I adjusted it to 0.06volts (a bias current of 60ma). This is within limits and should represent a safe level, although it is a trifle low if anything, resulting in less clean headroom for the amp, but that is OK for testing. This time, all was well, and extended playing at high volume had no detrimental effects, so I now pronounce this tube amplifier well and truly working!

After the chassis checked out, I spent some time figuring out how to mount it in the cab I bought. I finally settled on a scheme that used L brackets to screw the chassis to the back plate of the cab - I drilled the chassis appropriately and mounted it - all was well, even the knobs had enough clearance for full travel, something I was a little worried about.

With the chassis in the cab, my amp now looks like this:

In the final picture you can see the bolts I added to the back plate to bolt the chassis in - it fitted really nicely and I was pleased about that. As you can see, there are a couple of final cosmetic steps I need to take. Firstly, I’ll be ordering an etched plate for the front panel, and a nameplate saying “Cockburn A.C.1” or some such. And the innards are exposed out of the back vents - there are potentially lethal voltages accessible through those holes so I will get some metal grill or similar and staple it across the inside of the holes to prevent fingers and objects from going in there.

My thoughts are now turning to recording and a few mods. I think I will add an effects loop straight away, and consider some way of adding a switchable additional gain stage so that I can get a more high gain sound out if it if required. I will also experiment with compressors and boosters to see how that effects the sound.

All in all, a great days work!

January 6th 2008

I had some time this weekend and have started making a few minor modifications now that the basic amp is working, I also worked on some cosmetic touches.

First, I replaced the plastic input socket with a metal one, makes it look a lot more classy you know :) My next challenge is going to be to get the faceplate etched. I have no measurements for the chassis and didn’t feel like spending a lot of money for a laser etched faceplate that didn’t fit so I worked on making a dummy. I designed it in PaintShop Pro, and then since it was too big for a single sheet of paper, printed and laminated it at Kinkos. The result was a laminated faceplate that would give me a feel for the layout. I put it on the amp and was surprised at how good it looked. Here is a pic, also showing the new metal input socket.

I was pretty pleased but need to make a couple of tweaks before i send the faceplate order in - some of the holes need to be a touch bigger, and some of the lettering is hidden under the top of the amp - I’ll probably dispense with it.

Next i decided to tackle the effects loop. Rather than drilling another 2 holes in the chassis, I decided to do another mod first. I replaced the 3 speaker sockets with an impedance selector switch and a single socket. I had to drill for the switch but this let me free up 2 sockets for the effects loop. I studied the circuit and made the break just after the volume pot. All mods worked as expected, and I can now add delay, reverb and chorus to my hearts content :) I didn’t have any screened cable to make the runs for the effects loop and was a little worried that I would get some increased hum but it seems fine. I will probably do the job properly at a later date. That’s it for minor mods - I will use it for a while now, try micing it up and get the front plate sorted before I leap into more major surgery around a high gain mod.

Here is the chassis with the impedance switch and effects loop mod in place:

And that concludes my amp building journey!

END