SEL up and running in Canada
AX84.com | Mitch Markin | Nov 18th, 2012

I finished my SEL build a few weeks ago. It's a great sounding amp. I've been having a lot of fun with it! Too bad I'm building it for somebody else and I don't get to keep it.

It's a Doberman kit with the variable bias option and the optional Heyboer output transformer. It came with two JJ 12AX7's made in Slovakia and a Valve Art 6550 made in China. Chris shipped the kit promptly.

The documentation is excellent. The board and all the parts are first rate. The kit went together well. It took about four days to build, working mainly in the evenings, and it worked the first time.

I ordered the kit with the chassis unpunched and drilled the holes myself. I used a step bit for the tube socket holes.



The next step was installing all the pots, switches, jacks, tube sockets and transformers.




I sorted out the parts and wired up the eyelet board next.



Then it was time to go back to the chassis and wire the AC, the heaters, the input and output jacks and the output transformer secondary.

The transformers use stranded wire which tends to not stay where you put it, so I use cable lacing to keep the various bundles together. I think it looks neater than nylon cable ties. It's more of an old school military look and it's not that hard to do. I build Hiwatt clones, too, and that's how they did it.

I tested the AC wiring at this point. I made sure the secondary wires from the power transformer weren't touching anything, plugged in the power cord and switched it on. The pilot light came on. I checked the AC voltages coming out of the transformer and they were good. I checked the heater pins on the tube sockets, too. No tubes were in yet so they read about 7 volts.




All that was left now was installing the eyelet board and wiring it to the tube sockets and pots.




I checked all my wiring using the layout drawing. Then I used an ohmmeter and the schematic to make sure everything that's supposed to be connected actually was connected. I did a final check to make sure the polarity was right on all the electrolytic capacitors, then I fired it up.

There was no smoke so things were looking good. I measured the B+ voltages and they looked good. I shut it off and waited for the capacitors to discharge. Then I plugged in the preamp tubes and turned it back on. I checked the voltages again and things were still good. I shut the amp off again and waited for the capacitors to discharge. I plugged in the 6550 and a dummy load and switched the amp back on. Still no smoke and the voltages still read pretty close to what's shown on the schematic.

The amp has the variable bias option. I plugged my meter into jacks and adjusted the bias control so I read 80 mV on the bias test points. This means that the cathode current on the output tube is 80 mA.

The test points are across a 1 ohm resistor that's in series with the tube's cathode. Ohm's law says V=IR so the reading in mV is the same as the current in mA.

With 350 volts on the plate that means the dissipation is 28 watts. (Ohm's law again, P=VI) I could probably crank the bias higher with a 6550 but it sounds good the way it is.

Now it was finally time to plug in a guitar and hear how it sounds and it sounds fantastic! The cleans are outstanding and it gets good and raunchy when you crank it up.

I tried the amp with several other output tubes, a 6L6, a 6V6 and an EL34. The socket is wired as shown in the layout with pins 1, 7 and 8 all connected together then grounded through the bias network. I had to adjust the bias for each tube but they all worked well with this wiring.

There were differences in the tone with each tube. The 6550 and EL34 sounded similar. Both were bright and ballsy with the 6550 being louder. The 6L6 sounded darker but that could be because it was an older tube. The 6V6 was not as loud or bright as the 6550 or EL34 but I prefer the way it sounds. I put the 6550 back in, though.

I also tried both the 4k and 6k taps on the Heyboer output transformer's primary but didn't notice much difference. I went with the 6k tap.


I had the faceplates for the front and back made at a local laser engraving shop. The material is called lamicoid. It's white acryllic with a thin black layer on top. It's usually used for signs and name tags. It's similar to the traffolyte material that Hiwatt used for their faceplates and emblems.

The laser burns off the black layer where you want the lettering. This produces white letters on a black background. Mechanical engraving tools can be used, too. Also there are lots of other color combinations available.

I did the layout using Inkscape (free software similar to CorelDraw) which can output the artwork as a vector graphics file in EPS format. The engravers used this file to run their laser. The laser cut the holes for the switches, jacks and pots, too.

The person I'm building this amp for wants it in a cabinet made from solid rosewood. I bought the wood a couple of days ago so the next step is the woodwork. I'll be building two cabinets, a head cabinet for the amp and a separate cabinet for a 12" speaker.

This is a great little amp. I'm going to hate to give it up. I might have to build one for myself, although I have a P1-EX and my next build will be a 5E3 Tweed Deluxe clone. I'll be keeping those!

The bypass cap on the first stage is the what came with the kit. The BOM says it's .68uF at 25 volts. The way it's sitting I can't read anything that's written on it such as the voltage rating or the tolerance. I can't remember if it's 25 volts or higher but it is .68uF.

The two speaker jacks on my SEL are wired in parallel in case the user wants to run two cabinets. The 4, 8 and 16 ohm taps on the output transformer are wired to a 3 position rotary switch whose common terminal goes to both jacks.

I use the "running stitch" shown on the Naval and NASA sites most of the time. It's pretty easy and quick once you get the hang of it. Using waxed lacing cord makes it easier.

It took a while but the cabinets for the SEL and its speaker are finally done. Glad my customer is patient!

The cabinets are Solid Rosewood.

I built it as a head with a matching speaker cabinet. The speaker is a 10" Eminence Ramrod. The grill cloth is "Mesa style" black cane that I got from Mojotone. I got the handles and feet from Mojotone, too.

The rosewood was fairly easy to work with. It's like walnut but the dust isn't as irritating. It went through my thickness planer well. It's a bit chippy to cut though. I had to be very careful routing the dovetails to avoid tear-out.

Rosewood is somewhat oily, but not as bad as teak. I had to wipe the wood down with lacquer thinner before applying glue or lacquer to make sure there wasn't any oil on the surface that would affect the bond or the finish.

It cost about $200 for the wood to make these cabinets. I bought it rough sawed and had to plane it to finished thickness.

The cabinets are finished with five coats of lacquer. I used black hammertone paint on the back panels.

Here are some pictures:


Final Assembly


The Stack



Head



Speaker


It's pretty much a standard SEL but I'm running it with a 6V6 power tube. The customer doesn't want a loud amp and it sounds good with the 6V6. I've biased it at 38mA which runs the tube at about 12 watts. Maybe I should have just made him an HO, but this way if he wants more power later on we can just put the 6550 back in and crank up the bias.

The only other change I made was with the heater wiring. There was a lot of discussion here about hum a couple of months ago. This amp didn't really have a hum problem but there was a bit if both gains and the master were cranked up full.

I tried the suggestion of using an artificial center tap on the heater supply and that got rid of the hum completely. I tried connecting the center tap to ground as well as elevating it to the DC at the output tube's cathode. Both worked but it was a bit quieter with the center tap elevated.

Here's a picture of the mod.


I got rid of the link between pins 7 and 8 on the power tube and replaced it with a 100R resistor. I installed another 100R resistor between pins 2 and 8 to complete the artificial center tap. Pin 8 still connects to the bias network and it still connects to C6 via the purple wire. The green wire connects pin 1 to ground so the amp can work with an EL34 if desired.

I'll hate to see this amp go, but I'm already thinking about my next build. It will probably be a 4-4-0 Studio.

I used 18 gauge perforated metal with 3/16" holes for the grills. It's not hexagonal but the holes are arranged in a staggered pattern.