May/June 2016
Some notes and observations made while building a Tweed Deluxe 5E3. The opinions expressed are my own; use at your own risk.
Please do not copy and repost these photographs or link directly to them from my server.

I'd had a Hammond power transformer sitting around for years but for the life of me couldn't remember what I had intended to build with it. It seemed a waste to have it sitting on the shelf next to my also disused Marshall JTM-30 combo amplifier. The JTM-30 had that classic Marshall sound that suited the project I was working on when I bought it but has languished in a corner ever since. Suddenly the answer seemed so obvious - build the antithesis of the Marshall. Something crude, unrefined and undisciplined.

Looking back I think I may have been subliminally influenced by this shot I took at Guitar Center in Hollywood that shows two Tweeds surrounding the 1962 Princeton I was shooting.

The choices you face when you decide to make a classic Tweed Deluxe are bewildering - you can get everything from an inexpensive Chinese kit to something that costs what you might get a small used car for. But let's face it, the magic of the 5E3 is the circuit itself. Fender didn't select parts for "mojo" so there's no reason to do that now.

Since I already had a power transformer I settled on a kit from Tube Audio Supply that included the chassis and all parts except for the tubes and transformers. George was easy to deal with and included the output transformer for a reasonable fee. I was confident I had the most of the other parts to complete the build.

When the kit arrived I started to sort through the parts as usual and check to see if my Hammond was going to physically fit the chassis. The chassis itself looked incredibly sturdy, made with 16ga. steel that is spot welded at the corners. But while it looked like I would be able to make my Hammond fit I suddenly realized that it had no filament winding for the rectifier. So I never had a reason to go down this path in the first place.

I put the Hammond up for sale on Ebay and bought a 290BEX from Evatco. While their website recommended the 290AEX the footprint did not fit the existing mounting holes in the chassis. The BEX fits perfectly.

The turret board went together easily enough, but I noticed when I finished that components are commonly installed with leads going into the holes rather than wrapped around the turrets. My justification was to keep the holes for the underside wiring so that pushing a wire from the bottom did not risk pushing the component out from the top.

I could not work out how the board was supposed to mount in the chassis so I finally had to email George. He explained that I would have to decide where I wanted to mount it then drill holes in the turret board.

Here's what I ended up with (in inches, of course). I had some cloth wire hanging around from another build and decided to use that for the majority of the visible sections of the wiring. As you can see the yellow wire attaches to the underside but passes through a hole to the top.

With the underside wiring complete the board is taking shape.

Before you mount the board, however, make sure you mount the output transformer first or you will lose access to the nut that holds it into place.

Otherwise you'll have to do as I did and loosen the mounting screw for the board to access the output transformer nut.

I started the internal wiring with the heaters following Hoffman Amplifier's guidelines. Having encountered difficulties before with twisted cloth covered wire I used PVC insulated wire instead. For the remainder of the wiring I used Triode Electronic's PDF. When powered up the first time it delivered a mighty buzz regardless of knob position. The 290BEX has no center tap on the heater winding. Two 100 ohm resistors on the lamp terminals to ground fixed that issue.

To make the box joints for the corners I constructed a jig out of the mitre gauge that came with my table saw. Though the thumbscrew was tight, it slowly drifted away from square, ruining several boards in the process.

Because I did not want to ever concern myself with the mitre gauge again, I built a new jig that should never suffer the same problems. Rails run in both channels of the table saw ensuring that it will never drift too far from parallel to the blade.

My old dado blade was never very good, so I tried to see if I could use two identical normal blades with a spacer in between.

Unfortunately the gap was too great and left a strip of material in the center of the cut. I ordered a new dado blade on Ebay or you could try Carbatec.

Meanwhile I kept busy with the other cabinet details. I found a PDF from Modulus that I used in conjunction with the one from Scott Cole. I've marked out the radius for the top panel cutout so I can use the Jasper jig with a router to make the cut.

After making the radius cuts I used a straightedge to guide the router on the straight cuts.

Top rear panel marked for routing.

Half-moons cut.

Because the material is not very thick I had to use a flush cut bit to cut another layer of material for the roundover bit to ride upon.

The bottom rear panel radius is outside of the panel itself, necessitating a jig.

Glueing the first strip on the speaker baffle.

Solvent based black stain applied to speaker baffle.

Oxblood grille cloth and tweed tolex arrived from Evatco. Appearance is not too bad for vinyl.

Top of grill cloth clamped to table.

Grill cloth doubled over and stapled to edge of speaker baffle. I've seen the grill cloth attached to the back of the baffle in online tutorials but all photographs of original Tweed Deluxe show no sign of grill cloth from the rear.

Clamped grill cloth stretched tightly against the baffle. I tried using normal stainless steel screws glued into place but this clearly did not work. Tee nuts are unheard of in Australia so I ordered some M5 sized ones from Aliexpress.

One meter of tolex appears to be ample for the size of the cabinet. But orientation of the pattern is important to the overall appearance so some material may be wasted getting the orientation correct.

Laying out the tolex for the lower back panel.

Using the traditional DIY method of solvent based contact adhesive rather than Super 77 spray or water based adhesive.

Had a roll of copper tape lying around so used that for the rear panel shielding.

With the arrival of a new dado blade from the US I had high hopes of getting the box joints done. Though the ancient, cheap Ryobi table saw I have has an Imperial 5/8" shaft the bore of the Freud Dado was slightly less than that. After getting it stuck by forcing it onto the shaft I used some carbide sandpaper wrapped around a reaming tool to gently and slowly enlarge the bore.

A couple of threads came up in a search regarding this issue with the OPs being told that they had a problem with their shaft diameter or a build up of deposits on the shaft. That does not explain why my old dado set fits without any issue nor why the Freud, when measured with a caliper has an inside diameter less that 5/8".

During this time I started to worry about my Ryobi table saw. The mechanism that raises the blade failed. This ended up being an oval disc press fit onto a shaft that had worn. Fortunately the local Masters Home Improvement store had circlips that fit the shaft. This fixed the issue with the blade elevation.

That did not keep me from looking for a new saw. But most of the likely candidates here in Australia have metric shafts 30mm in diameter. That would render my new Freud dado blade useless. The Makita looked promising when a helpful staff member at Bunnings opened the box so I could have a look at it, but it appears that the factory modifies machines for other markets with a 25.4mm shaft - in other words one inch - while supplying a 5/8" shaft for the US.

Bosch apparently does something similar with some users in the UK reporting that they have successfully retrofitted US parts onto a nominally 30mm machine. The thought of spending the time, energy and money to convert these machines seemed too onerous. But then I noticed that DeWalt sells two models that they advertise as having "16mm" shafts. Upon inspection these were found to have 5/8" bore saw blades fitted at the factory. Considering I passed on a finished pine cabinet with baltic birch speaker baffle covered in the correct tweed fabric due to shipping costs from the US it seems crazy to spend over $1000 on a new table saw as well.

The new Freud blade gave a much nicer cut compared to the old one even on my decrepit Ryobi machine.

After doing a dry run I glued the joints and assembled the frame but remembered at the last moment that I had not cut the rabbet for the front facias.

Because I had cut 1/4" fingers for the box joints and the Modulus document erroneously(?) specifies 9.5mm for the rabbet I should have cut them short of the fingers. I say erroneously because a 9mm facia is specified. 9.5mm brings the facia 0.5mm short of the front angled edges. But my mistake of cutting the fingers will have the consequence of leaving a gap at the corner that will need to be filled.

Note the gap at the top right hand side.

Another site mentioned using a 3/8" roundover bit for the edges while the Telecaster forum shows that an original 5E3 cabinet had some variation but was nominally 15mm. I used a Carb-I-Tool 1/2" roundover bit. Note the gap at the top and bottom corners created by the rabbet.

Now I'm not so sure about covering the cabinet with the faux tweed vinyl. The box joints are not only incredibly strong but quite attractive as well. I've decided to seal the cabinet with solvent based varnish initially. This will not preclude using contact adhesive in the future to attach the vinyl. The varnish will help seal the pine against moisture. Scraps of the New Zealand pine that were left outside in the cold winter air developed a dramatic curvature.

Since the cleats will hold screws that secure the two back panels I decided to use Tasmanian Oak rather than pine for durability. At some point the top rear panel will surely come off to service the amplifier and the thought of gluing toothpicks in the screw holes to secure the screws once they have stripped out the threading sounds unattractive.

While waiting for the tee nuts to come in from China I assembled the amplifier and speaker in the cabinet to try it out. While it sounded even better than I expected it to, there was still some residual hum. The PVC covered wire did bother me so I decided to try cloth wire again, using some nice heater wire I still had in surplus. Finding again that it doesn't twist cleanly when using the drill method I instead twisted it by hand.

The heavy gauge solid core wired allowed me to get closer to the ideal I had in mind. This did reduce the hum, but there was still far too much. A search revealed several individuals posting what the Hoffman layout describes but without attribution to the source. Some even represented these layout and schematic modifications as their own opinions! The Triode Electronics document, as well as others like the Ceriatone layout show a grounding scheme different to the original and unfortunately flawed.

Because I was not using a Hoffman board I used a layout more closely following the turret board I was using. That was a mistake. Even an original Tweed Deluxe could be reworked to the Hoffman layout, likely decreasing residual hum. Now it is very quiet.

Even though I thought I was being very careful the varnished pine started to show scratches on a number of surfaces.

One distinctive characteristic of the Tweed Deluxe is the way the material overlaps near the corners. To facilitate this I masked the cabinet to leave a clean line of contact adhesive.

Then I cut the vinyl to the approximate size and marked each edge, then traced the cabinet's outline.

Here are my measurements for the right side.

After a disastrous application to the right side, I found that getting one edge down first then putting down approximately 1" widths down as the material is stretched produced an acceptable result.

Contact adhesive is unforgiving. You don't want to start with the vinyl misaligned. Taping the material to the front panel worked to keep the sides parallel and square to the cabinet.

If you slather on the contact adhesive onto both sides of an inside corner you'll have little luck getting a sharp inside corner. Mask off one side so the vinyl will sit flush to the corner.

Unless you do this sort of thing regularly it is difficult to picture how the material will conform to a radiused inside corner like this one.

Fill the split with a triangle of material.

Don't use lacquer thinner to remove excess contact adhesive or you will remove the pattern from the vinyl.

I used Palatino Linotype font for this reproduction tube chart.