Yeti’s First Amp – AO29 to Bandmaster

Yeti · 2018-08-29T19:44:57+00:00

Hi everyone, I’m starting out on my first valve amplifier build and thought I’d keep a running project log here to share my joys and (inevitable) frustratio...

Yeti’s First Amp – AO29 to Bandmaster

Yeti

Just a quick (and potentially not very interesting ?) update today.
I’ve cleared a bit of space and setup a decent sized work table. I’ve also had a scratch through my father’s (enormous) stash of components. His projects were typically op-amp based low voltage circuits so aside from a few resistors, switches and wires there isn’t much I can use.
I did manage to find his signal generator and oscilloscope though, which should help a lot for testing.

Finally, I got my hands on a 150W incandescent bulb for a test current limiter. It’s a little smaller than I’d prefer but I think I can use it for the initial startup, even if it isn’t good for setting bias... maybe I should try to find a second and wire them in parallel.

So yeah, looks like I’m ready to get going on the actual work… just as soon as I have a bit of time.

(I considered including a photo… but it would just be a photo of a table in a messy room with an oscilloscope on top of it so… yeah, maybe not ?)

This is more or less the plan from here.
- Fire up the old amp and take some voltage measurements to see if I can learn anything about the power transformer.
- Finish designing the power supply circuit and do a complete circuit review
- Start sourcing components (already have the hard to get stuff so should just be a matter of a couple of parts list orders)
- Start breaking down the old amp… looks like everything is riveted in place so this may be a long process.
- Start the actual build!

modulator

Rob Rob's site is gold. There is one thing he mentions, a treble removing cap over the feedback resistor, which might be handy for the single coil fans with sensitive ears. So many things to play with!

Yeti

modulator thanks for the tip, I'll check it out. I was hoping to use the tone cut to tame the ice pick but if its a simple as a cap I may look at that too.

Yeti

Fired up the AO66 to take a few measurements and decided to take some photos while it's out. Last hurrah in its current configuration. Next time it gets power it will have a lot fewer tubes!

Yeti

Having taken some voltage measurements on Friday night (more of that in a moment) I dedicated yesterday afternoon and evening to stripping the old amp. I tried to preserve as much that is directly usable as possible. This includes:
- The power transformer and its wiring
- The rectifier tube socket and its wiring
- The amplifier tube sockets (I did destroy the brown noval, which is a pity but I’m intending to use it right now, so I’m not devastated)
- The heater wiring to the sockets
-The output transformer and its connections to the power tube sockets

I’ve left some terminals attached to what will ultimately be the front face of the amp but I’m not sure if I’ll end up using them or stripping them at a later stage (they are riveted in place so stripping is a pain)
I’ve also left the expression pedal housing (silver box on the top) because it’s the same height as the power transformer and allows the amp to stand nicely on its head. I may even leave it there to protect the tubes until I build a proper cabinet for the amp.

Back to the voltage. It looks like the secondary of this transformer is +-375V, which is quite a lot higher than the +-335V on the schematic. This, coupled with the fact that it was originally meant to run through a tube rectifier and I’m intending to use diodes, means that my B+ is going to be high… up around 490VDC. Even putting this through a stoutish resistor will put it at the upper end of what the 6V6S can handle.

So… I have a couple of options. Either I can look at dropping the voltage… or I could upgrade the power tubes ?

Dropping the voltage could be as simple as putting an extra filter stage ahead of the power tubes but I might also look at just using the tube rectifier since I have it and it’s already wired.

Changing the power tubes to 6L6s is very tempting, but there are a few draw backs. Firstly, I want to play this at home and I’m worried they might just be too loud (time for that l-pad) secondly, I doubt that the poor output transformer would be able to take it. It was specced for 6V6s and I doubt that they would have included enough fat to move to their bigger brothers, even if I could sort out the impedances.

I’m going to toy with the options and see what I come up with. I'm hoping to finalise my design and start ordering parts in the next few days. With a little luck I might even be able to start the build proper next weekend.

modulator

The pictures of those glowing tubes are so nice. Did you try to play some sound through it at all before removing everything?

The "cleaned" chassis looks good. As for your high B+ possibilities, seeing as you are going to use diodes to rectify, is there a way to use a voltage regulator at all? I'm sure they do exist? I have been thinking about that route for filament supplies, in the far off future. If you look at my build, I started off with 470 V. That is now down to around 350 V, but your amp will require much fatter dropping resistors than mine. You have enough space on that chassis to include a second transformer just to lower the high voltage side of things. Might be the easiest to do, if you can find a suitable transformer. You could play with a bucking transformer on the mains supply side, and see where that takes your filament voltages. I have read online that some projects ended up with 6.9 VDC on the filaments, ouch. If you are there, lowering the mains side of things will help a lot.

Your amp will be too loud. Plan a second project, micro 5F1 1 Watt "home" amp? I need to start looking for components for that.

Yeti

modulator
I didn't get to put any sound though unfortunately. Partly because this is only part of the organ and things like volume controls, etc. aren't on this board so I'd have to figure out how i/o ports worked, partly because I didn't get the 12AX7 tube with the amp... but mostly because I didn't know where to plug the guitar in. ?

Thanks for the voltage suggestion, I'll take another look through your thread. My issue with such a big resistor is that it's likely to really pull down when at high volumes and I'm a little nervous of that. A buddy of mine suggested that I put in a voltage based power attenuator. This will serve the dual purpose of allowing the amp to be played at lower volumes. I haven't looked at the circuit yet but I'm sure there will be a way to set it up so that it limits the maximum voltage to something lower than B+.

I love the idea of that 1Watt amp. It's actually where I started before the scope creep kicked in. ?
Looking forward to following your project. ?

Yeti

Sorry for the gap between posts. There has been a little too much real life lately and I haven’t had as much time to work on this as I’d have liked.

I have refined my schematic a little bit and I think it’s more or less in its final state.
I’ve removed the effects loop (turns out they are a little more complex than I’d hoped) and the tone cut (got some advice that it really shouldn’t be necessary with my circuit).

I’ve also made a few additions:
- A ‘brownface’ tone stack switch, which adds an extra capacitor and resistor to simulate the 6G4-A Super-Amp circuit.
- A built in -9dB L-pad on the output
- A built in resistive dummy load

I’ve also changed to tube rectification from solid state. I already have the tube wired into the chassis and I have some voltage to shed so I thought I’d give it a try. It’s also relatively easy to change to solid state at a later stage if I need to. This is intended to be a test bed amp after all.

I was given the idea of using a power scaling circuit to further reduce the B+ voltage to a manageable level. I intend to use this to drop the voltage to 420V so that I can use the power circuit from a Deluxe, which also used 6V6s. I am using a fixed bias circuit for this amp so, unfortunately, I will not get the full benefits of the power scaling. Instead I’ll have to mount the pot inside the chassis so that it can’t be fiddled with. I will be able to move it between high and low power but any change would mean re-tuning the bias voltage so that’s not exactly a quick exercise.

I’m hoping to order the parts this week and with a bit of luck I’ll be able to make a start over the weekend. I’ll start with the chassis work, which will include removing the last few items that I don’t need, installing plates over the existing holes and of course drilling the holes for the new components. I’ve got a rough idea of how I’m going to lay it out, but I think that I’ll wait until I have parts in hand before I actually start cutting.
The plan after that is to wire everything point to point, partly because it’s supposed to give better tone/less noise but mostly so that I can play with the layout more easily going forward.

I’ll try to put together a rough sketch of my layout a little later in the week.

Yeti

Well, components have been ordered but it turns out I was a little optimistic on the lead times. ?
I’ve already received the first package but I’m not likely to receive much more before the weekend. Don’t think it’ll be a problem though because I have quite a lot of chassis work to get on with, including removing the last extraneous components, cleaning it thoroughly (possibly giving it a quick sand if I feel industrious) and of course drilling the holes for the new components.

I’ve put together a simple sketchup model to help plan the layout. I’ve tried as far as possible to keep everything transformer side of the old expression pedal tower (big light grey box… I’m keeping it there for the moment because it’s the same height as the power TRX so you can easily turn the chassis upside down). The broad idea is to build a second channel of some flavor in the remaining space at some indeterminate point in the future. My big constraint is the output transformer as I don’t really want any controls directly next to it.

The white cladding on the outside is intended to be 50mm aluminum bar that I picked up some time ago. There are holes from old components on all four sides, which need to be covered but I don’t want to lose space on the inside so I’m thinking of using it as face plate. The only downside is that it’s slightly narrower than chassis is tall so it won’t cover the full height. I’ll have a look over the weekend and see if it’ll work.

I’ve also included two terminal strips on the underside that should help quite a lot with the point to point wiring.

If all goes according to plan (and I don’t cut my hand off in the process) I should have a prepared chassis by the end of the weekend and I’ll be able to start on the component layout and assembly.

The 'black/brown' toggle in the tone stack switches between standard blackface and a pseudo brownface tonestack with the center position being 'raw' mode.

Yeti

I spent a very pleasant morning yesterday working on the chassis (and then a very pleasant afternoon at Garage Days Pretoria, thanks all) and have made quite a bit of progress. I have cut my aluminium stripping to size and cut out the various holes to mount it onto the chassis. I have also laid out the face plate and installed the output jack. I realized after the fact that I missed the attenuator/dummy load but as I don’t have the resistors yet it’s probably better if I hang fire for the moment.

Now I am by no means a metal worker and don’t really have the tools for precision cuts so some of it is a little rough, but I think it’ll do the job (missing screws are hard to install because of the transformer so I left them out for the moment). There also wasn’t any protection on the face of the aluminium (it is stock bar rather than face plate material after all) so it was almost impossible to avoid nicks and scratches. In trying to clean off a burr at one point I noticed that I could get a fairly decent brushed metal finish using a sanding disk on a baby grinder. I gave it a try on an offcut and the effect is quite pleasing, but it would need to be more regular to look neat. I’m just not sure if I’m brave enough to try it…
Another alternative would be to paint it but I quite like the aluminium.

Finishing aside I still need to drill the faceplate for the actual pots, add the attenuator and mount the terminal strips I’ll be using for the point to point wiring. So I’ve gone about as far as I can go for the moment. Hopefully the components will start arriving on Tuesday and then I’ll make some real progress.

Quick side thought. Does anyone have any recommendations for making control panel labels? Right now my best option is printing some out, laminating them and sticking them on with presstick ?
It’s never been my intention to give this a professional finish but I’d like it to look neat.

guidothepimmp

Looking good. For the control panel.. maybe a place that does silk screening? Or easier yet.. a place that does sign writing.. draw up the panels and ask thrm to print it on vinyl colour of your choice. Then simply stick it on. Or you could do similar on to a magnet. Qphoto does printing on magnets..

Yeti

guidothepimmp Hmmmm, hadn't thought of that. I'll look into it.

Yeti

w00t! The parts have started to arrive.
I picked up the tubes from Postnet yesterday, my RS Electronics order (more obscure caps, pots and switches) is supposed to be out for delivery today and I’ll be heading through to Communica to pick up the more mundane stuff later this morning. I had thought that I’d placed an order with them and would just be going to pay and pick up a pre-packed bag but it turns out that their online system is more of a directory service than a web store.

Assuming everything goes according to plan I should have everything I need this afternoon (except the inevitable items that I’ve missed or mis-ordered of course) and I’ll be able to start with some electronics this evening.

The plan is to start with the two more fiddly parts of the power supply: the power scaling (just used for the controlled voltage decrease) and the bias voltage circuits. I’m planning on building them both on a single piece of veroboard since I have it in abundance. It isn’t really the ideal product for the purpose as it is really designed for lower voltage DC but I’ve done some reading and I believe that I will be able to do it safely. It turns out that the trimmer pot that I am intending to use is only rated at 250V so I’ve changed the value of the series resistor from 100k to 1M2. This reduces the range of voltages I can produce but as I’m intending to use this to fix the voltage rather than as a true power scaling I’m not too worried about that.

Once this circuit is built I’ll decide how to mount it in the chassis, after which I’ll be able to decide how to mount the dummy load and attenuator resistors. I should also have the actual pots and switches so I’m hoping to do the final chassis work on Saturday and make some progress on wiring work after that. Of course I fully expect it to work perfectly with no hum the first time I switch it on so I should have a complete product in just over a week. ?

modulator

Brave... Waiting to hear how the power scaling and bias circuit works out.

Yeti

modulator Me too... I'll be energising them from behind the couch. ?

Small update. I received the order from RS and got everything from Communica. Brief audit and I see they made a mistake with 2 capacitors and included an extra set of zeners for some reason. Total cost is less than R10 but it'll mean another trip to get the right parts since the caps are part of the tone stack and I don't want to play with their values.

They also didn't have the wire wrapped cable I was looking for but I think that I've located some shielded twinflex I may be able to use.

I'll put up a few photos this evening and hopefully have some good news regarding the power circuits.

Yeti

I've done a quick layout of the circuit I'm planning to build tonight for those who are interested. Top is the Power scaling and bottom is bias circuit, they'll be on the same piece of board with the tracks thoroughly cut between them.

I've gone out of my way to make sure that the runs are as short as possible and that I never have adjacent lines in the circuit. I intend to cut the lines wherever they are not in use. I'll also be re-enforcing the current carrying tracks with solder.

At the moment I have all the terminals concentrated in the centre of the board but I may move them closer to their components where appropriate. I am also going to provide test points on the B1, Bias and GND but I haven't marked these.

Quick key: Grey is used track, purple is unused track, white is jumper wire across the top and red are terminals.

Yeti

Wow I’m properly out of practice with a soldering iron. ?
The circuit is built but it took forever. I may have gone a little overboard with stripping and reinforcing the tracks but veroboard isn’t really designed for these voltages, so I wanted to err on the side of caution.

I abandoned the idea of grouping the terminals in the center of the board but didn’t really think my placement through because some of the terminal mounting screws will butt up directly against side of the chassis. I may unsolder them and turn them around next time I get to work on this, but I suspect that I’ll just accept that I need to connect the wires before mounting the board and move on.

The pots are also a bit smaller than I had hoped and don’t have numbers marked on their faces, but I decided to use them anyway since this is likely to be a set once and then left in place. I suppose that's the hazard with shopping online.

Unfortunately I ran out of time so, aside from layout confirmations and continuity checking I haven’t done any testing.

On a different note, I’m battling to find a spot in that enormous chassis for my dummy load and attenuator resistors. The problem being that they need to be bolted directly to the chassis as a heatsink and the location that I had in mind for them isn’t going to provide the connection I'd hoped. I think that I might have to abandon the idea of building them in and build them into a separate box. That at least has the advantage of making them re-usable.

Tuckstir

guidothepimmp Nice.. no idea what any of that means.. but i dig it

same..... interested and want to attempt to build an amp in the distant future.. but me and electronics... eesh..

Yeti keep up the good work.. very interested to see the final results..

guidothepimmp

Nice.. no idea what any of that means.. but i dig it

Yeti

guidothepimmp Tuckstir
Short version: I haven't burnt the house down... but I'm still trying. ?

petercornell
Thanks for the suggestion, I hadn't thought of that. Will take a look this afternoon and see how it looks.

petercornell

Looking good Yeti, nice project.
As a suggestion, you could mount the dummy load and attenuator resistors on the top of the chassis, same side as the valves and drill some holes for the wires to go through, remember some grommets to protect the wires.

Peter