Hybrid Guitar Amp

[valvekits]

One rainy Sunday afternoon last year I had a play with National LM3875 audio power chip.
I like prototyping, but generally stay well clear of vero-board due to some unfortunate mishaps in my youth.
So, using Nationals' data sheet I constructed an amp, point to point fashion on the back of an old processor heat-sink with maybe half a dozen components - easy.
A quick amp test and I was quite astonished, not Hi-fi, but certainly not Lo-fi either. I later discovered that there is a whole world of this type of amp construction commonly referred to as Gainclones. However, I had no reason to take this any further at that time and so it was put away for another rainy day.

Now rainy days in June abounded, so then I wondered if I could make a guitar amp based on this chip, I seriously doubted it but thought I would give it a shot.
First I made a simple FET front end using a 2N3819 as I have plenty, but it needed a clean supply as it was only too willing to hum.
I was hoping to overdrive the LM3875, but it's a bit too clever for you to do that.
Any overdriving or signal shaping needs to be done before feeding to the amp because the chip has inbuilt safety; a bit like crossing the road with the lollipop lady - you have to behave yourself.

I then played about with a FET based Mu amp; lots of gain with a buffer and I was able to obtain quite a satisfactory sound. Indeed hindsight would allow me to say I could have stopped right there although I'm sure I could have messed some more to try and emulate a valve sound.
However, since valves are not something I'm short of, I thought why emulate when you can actually use one? In other words a hybrid amp, perish the thought but I do like to find out for myself and it's something I'd not previously considered.

Problem is, I'm usually geared up for working either with valves OR with solid-state so I was constrained by the contents of my junk-box and what was to hand, I used a small 120V : 24V toroid, reverse wired to step up for HT. I had to pay attention to my ECC83 load line and used 1N4007s in the cathodes to give me 0.6V bias with 88V on the anodes. This is probably not good for Hi-Fi but okay for a guitar amp, one half driving the other half. I arranged the ECC83 for 12V heaters, the supply being derived from an LM317 regulator.
I used an IRF 840 Mosfet as a source follower on the valve output to drive a tone-stack, this being fed into the LM3875 chip via a Master volume control.
Finally, I fitted a gain switch on the chip so it can be operated in two modes producing around 15W or a touch louder and grainier at around 40W.
The prototype cab is built from an Ikea kitchen door, some old fool (me) had a senior moment and drilled the holes for the handle in the wrong place.

How does it sound? It's very clean and hum free, it's never going to sound like a real valve amp, perhaps a bit too clean for most pro's but not bad for a bedroom practice amp.

Eddie

[Diabolical Artificer]

Wow! Thats something else. Thats inspired me to have a go; I have quite a few LM chips knocking about. Nice one Eddie . Andy.

[val33vo]

Why not just use valves all the way ? EL 84,S 6V6,S ( in push pull of course ) all spring to mindl, or ifyou want a bit more power EL34,s or 807,s However I do realise this would add a bit more weight to the unit

Val33vo

[jay_oldstuff]

And cost, This looks a grate little amp i'm going to have a play myself, in fact i have a little 15W practice amp in the cellar with good speaker and output ic but issues in the preamp circuit methinks a good candidate to add a valve or two to it.

Jay

[valvekits]

I had already constructed the amp when I picked up on Neon Indicators Russian Valves thread. Pity because I think there could be an opportunity to fit a couple in here and that would relieve the problem of having to obtain a big HT voltage.

Eddie