Another octal preamp build, this time a Deluxe!

I built an octal Princeton of sorts last year, which used various bits from the design of a few different early SE octal Fenders. It came out great and I still think it is one of the better sounding amps I've built.

A recent thread in the tube amp forum about a vintage 5A3 got me thinking about building a PP octal amp and I dug through my parts bin and found some bits and pieces that might work and whipped up this schematic:



I've used the basic bones of the 5A/B/C3 deluxes, but with a few alterations to make it more interesting. I'm using a 6SL7 instead of a 6SC7 for the input tube so that I can split the cathodes and have both a grid leak and a cathode biased input channel. I've also put a switch on the NFB so that I can have that both ways as well. Lastly I'm using a power transformer, field coil speaker and PP 6k6 transformer pulled from an old radio console. I figure that all together this should give me something like 8 watts of output and be a bit quieter than a 2x6v6 amp while covering the full range of octal deluxe tone options.

 

Finally starting on this project!

I started building the cab last week. Although this is a Fender inspired amp, I'm going with something a little more Gibson like for the speaker cabinet. Here are my plans:



And the beginning of the cabinet:





And the template for the grille laid out on some MDF:

 

I also started working on the electronics. I built my chassis from scratch using a sheet of 5052 aluminium. This is my second time bending my own chassis and I think this one came out much better than my last attempt, this grade of aluminum was much easier to work with than the 6061 I used last time. I am especially happy with how the sides turned out and my decision to attach them with rivets. I didn't think to take any pictures of the chassis build process, but here is a shot of the finished product with the hardware installed:



As for wiring it up, I've gotten as far as wiring and testing the power section:

 

I've been making a little progress here and there 30 minutes at a time in the evenings but it's added up to the point that everything except for the control wiring and heaters is now done.

 

Finished! (Wiring anyway)

 

I had some motorboating when I powered it up for the first time with all the tubes. I rechecked my connections and found no wiring errors or bad joint and eventually decided it was probably due to insufficient power supply decoupling between stages.

I was surprised I had any issues since I lifted almost everything in the circuit directly from the octal Fender designs, but I realized that in adapting the power supply to my salvaged radio parts (which include a field coil speaker) and in the process of lowering the B+ to a range the 6k6 power tubes would be comfortable with, I ended up with much smaller dropping resistors in the power supply. It seems this left me without enough decoupling between the stages, particularly the PI and input tube stages.

I ended up increasing the value of the dropping resistor after the screen grid node and adding and additional filter cap node to separate the PI from the input stages. This took care of the issue and I managed to cram the additional power supply parts in there without it becoming too awkward.



And here's what the finished (except for labels) chassis looks like:



Next I need to finish building the cabinet. I played through it briefly and it sounds promising, but I was just using a disembodied speaker sitting on the bench next to the chassis so its hard to really judge. The unfortunate thing is I can't just run it through another speaker in a cabinet since it needs the speaker field coil and output transformer which are both part of the speaker, so I'll have to finish the cab before I can really play through it.

 

Thanks!

I am using the harbor freight brake, I have the 30" model. For this chassis, I used 0.062" 5052 aluminum sheet from Mcmaster Carr: https://www.mcmaster.com/catalog/125/3824 I believe that translates to 16ga. Make sure you take note of the alloy when choosing aluminum sheet, I used 6061 on my last chassis and while it is really easy to drill, it is almost impossible to bend using the HF brake. The 5052 I used this time was very easy to bend by comparison and it drills out fine (maybe requires a little more grinding to clean up the holes). It is reasonably strong, though if I were making anything with a bigger chassis size or heavier transformers I might go up to 14ga equivalent. I was worried that might be too hard to bend but this was easy enough that I don't think it would be a problem.

I am really happy with the rivet approach since you can paint over them and that makes it look more professional. I used the cheapest riveter from HF and it works beautifully for this.

The paint is just gloss rustoleum. No clear coat either and it has a nice high gloss to it. All I did was give the chassis a quick sanding and a single coat with the rattle can and the finish comes out surprisingly perfect... I just wish guitar finishes were that easy!

 

I'm still making slow progress on this, got the cab pretty much finished:



Next I need to go shopping for material to cover it, I'm planning to do a dark brown faux leather of some sort, in the style of the older Gibson amps.

 

I got everything installed in the cabinet and got to actually play it a bit tonight!





It sounds awesome. I'm really happy with how nice the speaker sounds, I was optimistic since it's got a ribbed cone, but you never know quite what your going to get with an old radio pull speaker. I'm happy to report that this one seems to suit the amp perfectly.

Overall I'm very pleased, but still have a couple of things to tweak. The NFB is very light and I realized I needed to check the speaker impedance (which I had avoided since the speaker is soldered directly to the OT). When I actually desoldered the speaker leads and checked I found out that I've got a 2ohm speaker. Since the original was designed for 8 ohm and I'm using lower output 6k6 tubes i figure I've only got something like 1/3 - 1/4 the NFB that was intended with the 5c3, so I'm going to put in a smaller NFB resistor and see how that sounds.

I've also got too much brightness on the cathode biased channel as it's the one with the bright cap on it. Normally the "bright" channel being bright would be a good thing, but it's too much, especially since I really intended the difference between the channels to be the input stage bias rather than brightness. I think I need to try a smaller cap value there, maybe 100pf or so?

I've still got to figure out my covering material too. Hopefully I'll get a chance to go shopping for some this weekend.

 

I need a sanity check, does anyone see a problem in my schematic that would keep the negative feedback from working? I've lowered the series resistor value to 100k which according to my calculations should give me let close to the same level of NFB as the 5c3, but I'm getting no effect at all when turn it on. It sounds exactly the same with the negative feedback on or off.

I've double-checked by wiring and I've got continuity between the transformer secondary and the grid of the tube so at this point I don't think there's any physical problem but I can't figure out why I'm getting no negative feedback effect, no lowering of volume or anything else.

At this point I can't think of any reason it's not working unless it has something to do with the field coil speaker but I don't see how that should cause an issue...

 

So, I figured this out about 5 minutes after posting the above plea for help... The 5 pin plug connecting the speaker and OT to the chassis has 4 connections on it: electromagnet send, EM return/OT center tap, and the 2 ends of the OT primary that connect to the power tube plates. If you are paying attention, you will notice there are no connections to the secondary side of the OT, because it is mounted on the speaker basket and wired to the speaker directly. So, I had to add a wire running from one side of the OT secondary back to the chassis to provide the NFB signal, but what I didn't think about was that there is no ground reference on the other side of the OT secondary and apparently that is required for the NFB to work.

After this occurred to me, I clipped in a test lead from the non NFB side of the OT to a bolt on top of the chassis and it finally worked! It is still a fairly low level of NFB, but the volume drops slightly and the distortion tightens up a bit when it is connected.

Now I just need to play with the bright cap values a bit to try to find a decent middle ground that works for both single coils and humbuckers and I'll call it finished.

 

Last weekend I finally got around to getting the supplies I needed to finish up the cab. I got a 1/8" end mill bit so I could route a channel for the piping:



I got some leather upholstery material from the fabric store (on clearance for $10/yard!).

I also needed beading/piping material and I didn't want to place a special order from Mojotone or someone just for that and I was also disappointed with the color selection they offer, so I looked around at the hobby store and found some leather cord meant for jewelry making that seemed like it would fit the bill. I got this variety pack for $3.99 with three metallic colors, each long enough to do two full wraps around the cabinet:



I was thinking I'd go with the darker gold, but once I got the cabinet covered, I decided it needed more contrast so I went with the lighter gold.



Here's what it looks like now with everything except for the grille cloth done:



The covering material is actual leather of some sort, which is kind of cool because the amp has a surprisingly supple feel like the interior of a luxury car. The down side is that the seams at the corners show more than I like because of the much lighter color of the back side of the leather. Its not all that noticeable in these pictures, but it is very apparent in person, so I am going to have to get some leather touchup paint or something to darken those areas.

 

It does ride on top, only partially down in the groove, but there is a seam there, I basically used the piping to cover the seam between the side pieces and the middle. This material is a bit thicker than tolex or Tweed so it doesn't quite leave room for everything to fold down into the groove. I think with tolex you'd need to use a slightly thicker cord for the piping, this stuff was 1.5mm so maybe 2 or 3mm would be good for a 1/8" groove and tolex?