My new amp design- video demo

[Steve Conner]

Hi all,

YouTube - Ninja Corvette Hybrid guitar amp demo

Higher quality version of the soundtrack (The soundtrack on the video is heavily compressed, both in the dynamic range sense and the codec sense. This was recorded at the same time using a handy digital recorder.)

More information here: scopeblog » Ninja Corvette Hybrid

I offer this new design in the hope of causing some debate. It's a 3 watt tube amp that also contains a frequency compensated attenuator, a roughly 30 watt transistor output stage somewhat like the old Thomas Vox/Leslie ones, and a "VVR" type power scaling circuit.

I've been working on it for years and have kept the output stage secret up until now. I still have some problems with DC offset when it's heavily overdriven, but as soon as I get that sorted, I will publish the schematics if anyone is interested. (It could be as simple as just AC coupling the speaker.)

I have a few other threads on the design, but I called it different names and am having trouble finding them.

[octal]

I thought ninjas were supposed to be silent? Might want to have a bit of a re-think on that name, mate!

Nathan

[dodona]

not bad at all.

[mooreamps]

Steve,

Very, very nice.... I really think you on to something here.. Also, kudo's for internal layout and such....

-g

[tubeswell]

Coool

[dai h.]

Neat!

[trem]

Wow.....You are a REAL Scotsman. I have read many of your postings on the forum Steve, and appreciate your input. You remind me of another sharp, amp guy that lives in Scotland, John Phillips. Though, I believe John was a transplant to Scotland.
I would be anxious to see what you have cooked up over there.
Best
http://www.youtube.com/watch?v=DZigVupYve4

[imaradiostar]

that's pretty neat!

Have you thought about making the wattage switch footswitchable?

It seems like it cleans up nicely with the volume down- pretty cool!

jamie

[Steve Conner]

Hi guys, thanks for the kind comments!

To address some of your questions here and on Youtube:

The wattage switch is 4-pole, 3-position. It would need a bunch of relays to make it footswitchable. If I built this amp again, I think I'd do it as a 2-channel head, with two completely separate sets of controls. Or, it's compact and cheap enough that you can just have a wall of them like Brian May.

I couldn't figure out how to add an effects loop between the tube and SS parts. The SS output stage is only in use on the high-power setting, anyway. It's not a "Herzog": the speaker winding of the OPT drives the bases of the output transistors directly, like those old transformer-driven Vox Super Beatles and so on. The only "silicooties" in the signal path are the output transistors and a couple of diodes, and there's no NFB at all.

[pdf64]

That's an excellent project - well done!
And the uncompressed soundtrack demo amply illustrates the '10 times the power to get twice the sound level' rule.
Is the 'dc offset at the speaker' issue due to there being no NFB?

[kg]

so you just hung some followers off the OPT secondary for the "high power" mode?

sounds very transparent...

[MWJB]

Very cool. I have been having similar thoughts (concept-wise), but utilizing low power twin triodes, rather than SE power section. Scaling the very low power output by physically subbing 12AT for 12BH7 in a self split PP arrangement, then a more traditional speaker driven line out for feeding additional power stages...not quite as compact, subtle & easily switchable as your design...ingenious!

[Mike Sulzer]

Hi Steve.

Does the transistor amp clip, or are you feeding the output of the tube amp into it with the level set so that the transistor amp remains linear?

[Steve Conner]

Hi Mike,

I demonstrate it both ways. When I first show the high power setting, I start out with the transistor amp linear, and then when I turn all of the knobs up full, it gets overdriven somewhat.

Of course, if transistor clipping was really that objectionable, then you would have noticed this straight away and wouldn't have had to ask.

One of my prototypes had +/-55V rails, which was enough that the output stage wouldn't clip under any conditions, but I reduced the rails to +/-35V for the version I showed here, since I discovered that the clipping sounded quite acceptable, and I thought with 20 less volts the transistors and speaker would be in less danger of destruction.

[mooreamps]

Too bad you live so far away.. I would have enjoyed conferring with you on some of my own projects....

-g