807 power amp #1
 

Its 6.08am now and i'v been sitting in my shed for like 6 hrs, building this.

Usually, i spend this part of the day watching Cops/Cheaters on channle 151 with our drummer. However, today channle 151 have become very annoying and did a whole day of stuff that wev seen before. So a new activity was needed.

I thought, even tho i'v got a number of 807's, and i'v lost interest in them, its worth at least building a prototype amp to see if 1) i'm capable of building a power amp of considerable power and 2) to actually see what the 807's are like out of curiocity.

So i built this power amp, and to be honest it aint too bad. Sounds nice, and 807's dont look TOO bad in the dark. In fact, these 807's seem to have a slight blue glow in the glass, which makes it look even nicer.

I'm using a 5Z4 rect at the moment, which is abit small i think but works. For the smoothign choke i used the primery of a mains transformer form a transistor amp, which seems to be quiet effective.

This also happens to be the 1st time i'v ever built a long tailed pair phase splitter. I used a 6SN7 and it worked 1st time no pronblem. I think the trick was to draw the circuit in a way that seemed logical to me. In schematics its usually not how my brain would consider logical. But...as soon as i redrew it i was like OOOOOOH i see and now i can build them with no problem, and feels like i'v known it all my life :).

I gave it a blast with my minidisk, and these 807's really sound ok. I doubt anyone appreciates listening to black metal at 6am, but they should be used to it by now.

So, the 807 isnt such a bad valve after all. And looks like i AM capable of building one of these. Even when slightly tired.

The search for an even cooler power amp valve still continues tho.

I will put a pic and a schematic up in the ''morning'' (3pm).

Re: 807 power amp #1
 

Ok

Here it is.

Re: 807 power amp #1
 

Hi Adi,

your phase splitter doesn't look right to me - as far as I can tell, yyou're driving both grids in phase (at the splitter and the ouput). I think on of the phase splitter grids should be connected to a fixed bias potential, with no signal.

Jim.

Re: 807 power amp #1
 

Yes - it does look a bit strange at first sight.

The two halves on the 6SN7 form a long-tailed pair with the grid of the second triode at ac ground. That's fair enough, but I would have expected to see the 1.2M R going from triode 1 grid to deck, not to the 0.47uF cap., and another 1.2M R from the grid of the second triode to ground and this R in parallel with the 0.47 uF.
That way, the second triode acts as a grounded-grid phase follower and the first triode as a phase inverter.
I can see that as it stands it sorta works - but I'm convinced that it ain't quite right!:-/

Al.

Re: 807 power amp #1
 

The phase splitter as drawn looks peculiar because it does not have any DC return path for the 6SN7 grids, they are just floating.

If the coupling capacitor between the EBC33 anode and the 6SN7 grid is removed and replaced with a direct connection, the problem is solved.

This minor change gives a phase splitter circuit that is fundamentally similar to the famous Mullard 20W amplifier or indeed the Mullard 5/10.

Ron

Re: 807 power amp #1
 

Looks like you're stuck in NZ time!

Re: 807 power amp #1
 

Yeah, the phase splitter is copied from some kind of hifi amp design i think. Although the original design had no dc blocker cap, as you said, i stuck one in. I thought if it doesnt work right i can allways remove it later.

Re: 807 power amp #1
 

Tut, tut! You wouldn't be the phantom core-twiddler would you? ;D

Best wishes with the amp. :thumbsup:

Al

Re: 807 power amp #1
 

At low frequencies (<500Hz), there will be an in phase signal passed to both ouput grids, and that is not good! (if you don't believe me, do the maths - the component on one grid is very small, but it will cause some distortion). The circuit you have is a frequency dependant potential divider, so long as the capacitor is large, the effect won't be noticeable, but if it was less than the current value, it would. If you want to stick with this circuit, increase the cap to say 1uF, or even more

You should seperate the AC ground of the two grids really.

Jim.

Re: 807 power amp #1
 

Jim

I make the reactance of a 0.47uF capacitor equal to 677 ohms at 500Hz. This is not really comparable with the phase splitter 1M2 inter-grid resistor, so phase shifts due to the CR network at this frequency are small (<<1 degree).

Even at 50Hz, its reactance is 6k7 ohms, which is several orders of magnitude smaller than 1M2 and the phase shift of the network is less than a degree.

Ron

Re: 807 power amp #1
 

Adi, that LTP is meant to be DC coupled, you can't just jam a cap on the input and hope for the best. As it stands, your LTP grids are at 0V while the cathodes are at 100V or something- it's biased WAY cold!

Can't you just copy the LTP from a Fender/Marshall or something? Basically you need two extra resistors, a bias resistor and another grid leak, so that each 6SN7 triode has its own grid leak.

Re: 807 power amp #1
 

Yes, but it is still there, and for the sake of a couple of components, it can be removed. As I said, with 0.47uF it works, but with a smalller capacitor, you could run into problems.

Having said that, with the music Adi plays, he may prefer to put in a smaller cap;D

Jim.

Re: 807 power amp #1
 

Thanks for the advice. I will take the cap out and just connect the grid directly.

I think overall the prototype went quiet well tho. It worked 1st time, which i cool.

I didnt have any 807 sockets so i just cable tied some wires to the pins and drilled holes in the wood for the pins to go into.

I thinki will try out some more power valves and then finally start buildign the real thing.

Re: 807 power amp #1
 

Excellent! Now that's what I call bread-boarding! ;D

Al.

Re: 807 power amp #1
 

I'm pretty sure thats a sort of acceptable method of doing it. I made friends with the 83 like that, and i tried out directly heated triodes with AC/pens and stuff liek that too. It seems to work for prototyping. Plus the fact that i prototype stuff on a piece of plywood with nails as terminals makes it easier. The wires are soldered to nails round the ''socket''. That keeps everythgin stable.