Omni Box - One Box to Rule Them All

[Ben N]

I hear that re bending the legs, I was more thinking of putting the socket sideways, but I haven't measured the clearance. 250v caps on a 200v supply doesn't worry you?

[vigilante397]

Not really, nope. I used to always use 400V caps with my 200V builds, but eventually got tired of spending extra money every build so I did some tests and found that with a stiffly regulated wall wart into this nixie supply set at 200V I had zero voltage spikes in about 200 hours. That was good enough for me, so I use 250V caps now.

All that being said, for designs I do at my day job if I don't use caps rated for at least double the expected voltage I get scolded

[anotherjim]

Very nice work.

This is now one of the clearest schematics you will find out there.
A few things I notice - which are trivial...

Original copied Fenders Hi/Lo twin switched jack inputs. When only the Hi impedance input is used, the x2 68k resistors are in parallel. That means you have 34k going into the first tube grid for which 33k is what I'd use with a single input, but I'm not sure anyone is going to hear much difference.

Treble pot can be linear; since it is really a pan between all treble via C1 and the bass, which is the full signal via C3 but with the highs bled off via C5.

The bass pot has to be log or all the control seems to be in the first 20% rotation.

The mids pot is really only controlling the strength of the bass & treble filters. 10k is original, but some may find it subtle. Some make the mids pot 250k which at maximum mostly defeats the tone controls and provides more drive to the second triode. I've fitted 10k, but I think I'd raise it around 50k or so in future.

The bright switch is very subtle with 120pF. Obviously, it never works anyway with the volume control at max. With the control around mid resistance, the source impedance into the next grid can be raised around 250k, meaning the tube Millar capacitance can audibly cut the highs. The bright cap corrects for this. For a more obvious bright, I like 470pF better, which is what Ampeg used in the same role.
My switch is an on-off-on. One way it connects to the volume wiper for Bright, the other way, it grounds the cap making it a Deep switch without taking any extra panel space. The extra Deep setting is useful for bass guitar.

I found the output far too hot as it is. So the final output off the 100n C2 is a 220k dropper to an A47k pot. The next stage in mine is an op-amp balanced output driver, which don't sound nice if clipping.

[vigilante397]

Original copied Fenders Hi/Lo twin switched jack inputs. When only the Hi impedance input is used, the x2 68k resistors are in parallel. That means you have 34k going into the first tube grid for which 33k is what I'd use with a single input, but I'm not sure anyone is going to hear much difference.

That would be a simple thing to change, I'll give it a try to see if I hear a difference.

Treble pot can be linear; since it is really a pan between all treble via C1 and the bass, which is the full signal via C3 but with the highs bled off via C5.

The bass pot has to be log or all the control seems to be in the first 20% rotation.

The mids pot is really only controlling the strength of the bass & treble filters. 10k is original, but some may find it subtle. Some make the mids pot 250k which at maximum mostly defeats the tone controls and provides more drive to the second triode. I've fitted 10k, but I think I'd raise it around 50k or so in future.

I have read a few posts in various places about people changing the tone stack, and it makes sense, but to be honest my ears like it the way it is, so I'll likely keep it the way it is.

The bright switch is very subtle with 120pF. Obviously, it never works anyway with the volume control at max. With the control around mid resistance, the source impedance into the next grid can be raised around 250k, meaning the tube Millar capacitance can audibly cut the highs. The bright cap corrects for this. For a more obvious bright, I like 470pF better, which is what Ampeg used in the same role.
My switch is an on-off-on. One way it connects to the volume wiper for Bright, the other way, it grounds the cap making it a Deep switch without taking any extra panel space. The extra Deep setting is useful for bass guitar.

That's an interesting idea, I may try that out if I ever do a similar build and want a bright/deep switch. Thanks That being said, with the current range of the treble control I'm considering leaving the bright switch off. Haven't decided for sure though.

I found the output far too hot as it is. So the final output off the 100n C2 is a 220k dropper to an A47k pot. The next stage in mine is an op-amp balanced output driver, which don't sound nice if clipping.

I agree. It goes fine into my discrete cabsim (which I prototyped last night and am very happy with tonally) without horribly distorting it, but it's still hitting my headphone amp way too hard. I still haven't decided whether I want to keep a master volume control or if I want to find a maximum usable level and replace it with fixed resistors. Now that I have a working cabsim I can run it into a direct out to see what a recording interface or PA system will see to determine what I need to do to the output level.

One thing I did terribly wrong that I didn't post in the schematic, when I built the preamp board I'm using for testing right now I attached a passive cabsim to the output, then right before I milled it I added a non-cabsim output straight off the pot and that's what I've been using. What I failed to think about was the fact that my signal still had a path to ground that included a voltage divider and a low-pass filter. So now that I pulled that off I'm getting an even hotter output and more treble response, and I'm also finding the master volume more useful (thus my tentative decision to get rid of the bright switch).

Thanks for the input Jim You've added even more things to my list of things to try out

[PRR]

> The mids pot is really only controlling the strength of the bass & treble filters. 10k is original, but some may find it subtle. Some make the mids pot 250k which at maximum mostly defeats the tone controls and provides more drive to the second triode. I've fitted 10k, but I think I'd raise it around 50k or so in future.

Where Fender fitted a fixed resistor instead of a knob, he used 6.8K in big amps and 15K in small amps. So if ~~10K is the "right fixed value", we could use a 100K 10% Audio taper and get the same effect around "5" (4 to 5.5), yet still have significant rise at "10". For Max midrange a switch to break the mid pot ground is simple and effective.

[Ben N]

For Max midrange a switch to break the mid pot ground is simple and effective.

...max midrange and level--you'll def need the MV then.

[vigilante397]

Little bit of progress tonight. I had finished the speaker sim last night and had added header pins so I could connect everything with jumpers for convenience. Tonight I added the DI transformer and played around a bit with recording. I had a special request on youtube to try it out with bass, so I shot a quick video with my bass. I'm pretty excited about how this sounds for bass, I definitely understand why people record bass with it. Anyway, here's a picture of what the messy prototype looks like right now:



And if you thought I was bad at guitar, wait 'till you see me play bass!

[stringsthings]

This project sounds great!  I'd definitely be up for a PCB.  SMD or thru-hole.  I"m not picky 

[vigilante397]

This project sounds great!  I'd definitely be up for a PCB.  SMD or thru-hole.  I"m not picky 

Thanks I'll put you on the list (which so far is just you ). My main focus is an SMD board that will fit in a 1590B, though I'm also considering doing a through-hole board that would go in a 1590BB, as a 1590B would be a bit of a squeeze. The tricky thing about SMD is that you're pretty much using the exact same parts I'm using to match my footprints, whereas TH gives you a little bit of flexibility, as long as you can make the pins fit it works.

[vigilante397]

So this project now has a deadline. My wife is 38 weeks pregnant, and I need to get the design finalized and order PCBs before the baby is born. I have a thru-hole cabsim finished but I want to verify my SMD cabsim layout, and I still need to test my new headphone amplifier. Once all those are finished I will get boards ordered and hopefully have a finished, working prototype by the end of the summer (assuming my wife still lets me build stuff after the baby is around).

[EBK]

So this project now has a deadline. My wife is 38 weeks pregnant, and I need to get the design finalized and order PCBs before the baby is born.

Better hurry!  Both of my kids popped out slightly earlier than expected, ruining whatever final-prep-before-baby-arrives plans I had made.   

[vigilante397]

It's true, but my wife has a pretty good track record so far. Our first was born on the due date and our second I think was only a day or so early. But as they say, every kid is different, so I will definitely be hurrying

[bluebunny]

My wife is 38 weeks pregnant

Will there be a thread for this project too?   

[bluebunny]

so I will definitely be hurrying

Yes please.  I've ordered some 6N17Bs...   

[anotherjim]

Those Russian submini tubes seem to completely bypass the UK. Where can you get them from? I can only find NOS of low gain US/Euro types, although they are quite cheap as such things go.

I built my F2 as a 2 channel 1U rack for "stereo bass". Used it with a PJ bass at a party recently. Had a "synth mod" DS1 and CryBaby on the neck pickup and straight deep bass off the neck. Did some Prodigy songs with just me and the drummer. Separate amplification is the key to stereo bass, as anyone will tell you.
Which is a long way of saying, don't just build one of them.

....and keyboards can use them too.

[bluebunny]

Those Russian submini tubes seem to completely bypass the UK. Where can you get them from?

Found mine on the 'Bay - Ukraine or Russia, I think...

[vigilante397]

Okay, headphone amp has now been verified. I went with a fairly simple NE5532 design and I really like the sound of it. It's not as high gain (amplification of 4.5) as the LM386 (min amplification of 20) so it can get nice and loud without overdriving. It responds very nicely to the master gain of the preamp, so I won't be needing an additional knob just for the headphone volume, and I can lock this thing down at 5 knobs.



My original plan was to build a thru-hole version of a couple circuits to verify the sound, then build an SMD version of the same circuits to more or less verify the footprints of my components, but now I'm thinking I might just triple check my pinouts and assume my layout is okay so I can get the boards ordered right now The only component I've never used before is my JFET, but now that I think about it I can just solder leads on it and try it out in my through-hole circuit, rather than have to mill out another board and get it all stuffed just to test one component.

I built my F2 as a 2 channel 1U rack for "stereo bass"

I have plans to eventually add more features and use a transformer instead of the SMPS then throw it into a rack, but that will be a ways down the road. There's a guy on youtube that keeps throwing ideas at me to add to this, and they're all really good ideas, but they're beyond the scope of what I planned for this project. I have an empty 1U chassis on a shelf in the garage waiting for when I get to it though

[vigilante397]

Okay, I got the headphone amp added on and I think my layout is finished. I'm still going to double and triple check everything before I order boards, but I think this is likely how it's going to stay. I also updated some 3D models I either had missing or had roughly drawn in myself so I now have a more accurate model of what it's going to look like fully assembled. Final board dimensions are still at 2.095" x 1.910" (53.213mm x 48.514mm) so I'm still planning on this being in a 1590B.







EDIT: Aaaaaannnnnd that layout is no longer accurate. I changed the cathode bypass caps (C16 and C17) to tantalums instead of aluminum electrolytics. I had already made everything fit with the aluminum electrolytics so it wasn't really about saving space, but I just like the look of tantalums

[vigilante397]

You know what, screw it. I've read my datasheets a dozen times, I've double, triple, and in some cases quadruple checked my pinouts.



Boards have been ordered. Paid extra for the 2-3 business DHL delivery. I should hopefully have my boards in about a week

[vigilante397]

IT'S FINALLY DONE!!!!!!!!

I already posted this in the pictures thread, and I plan on cross-posting it in the SMD thread as well, because I'm excited! The Boards from Elecrow came the day after my surgery on my hand, which was a bit cruel on the universe's part. I waited a little while but couldn't wait forever, and it took me about 4 hours to solder the thing up, because holding tweezers between index and middle fingers is hard So it's also pretty sloppy if you look closely at the board, so please don't look closely at the board